The effects of environmental factors associated with childhood urbanicity on brain structure and cognition

Urbanization has been progressing worldwide for over a century. In contrast to rural settings, cities attract people by offering opportunities in terms of income, education, health and social services [1]. Although urban areas may yield a higher quality of life on average, residing in cities may also have some negative impacts on physical well-being and mental health (World Health Organization and UN [2]. Previous studies in developed countries have shown that urban upbringing is an environmental risk factor for some psychiatric disorders. For example, the incidence and prevalence rates of schizophrenia seem to increase with increasing urbanicity [3, 4]. Studies in developed countries have shown that the gray matter volume (GMV) of the right dorsolateral prefrontal cortex is negatively correlated with urban upbringing [5]. In addition, in a social stress task study, researchers found that currently living in a city was associated with increased amygdala activity, whereas an urban upbringing affected the perigenual anterior cingulate cortex (pACC) [6]. These works suggest that the childhood residential environment might differentially influence the development of the brain, and early-life urbanicity is a risk factor for impaired brain development and mental health.

However, unlike the pastoral rural lifestyle observed in developed countries, inequalities between rural and urban settings as well as within urban areas have been persistent features in many developing countries [7, 8]. Moreover, inequalities in areas such as education resources, health care, housing and retirement benefits are observed in rural areas [9]. The differences in schizophrenia prevalence in China may reflect such issues; this pattern has shifted from higher prevalence in urban areas in the 1980s and 1990s [10, 11] to less apparent urban‒rural differences in recent years [12‐15]. In fact, the prevalence of mental disorders was higher in rural communities than in urban communities in China in the most recent national survey [16]. A recent study found that urbanicity was associated with perspective-taking and depression symptoms, and this was mediated by neural variables [17]. While early-life urbanicity may have benefits, it is important to study the impacts of early-life urbanicity on brain development and mental health in developing countries, such as China.

China has undergone large-scale urbanization since the 1980s as well as unique economic development [18]. Its’ urbanization rate has risen from 15–20% of the population to 51% in 2011 [19]. This provides a unique opportunity to study the effect of childhood urbanization. We selected subjects whose current urbanicity conditions were similar, thus maximizing the effect of urbanicity on upbringing. In our previous report, we analyzed voxel-based morphometry (VBM) data after adjusting for the total GMV, similar to previous research [5]. It has been shown in previous work that the total GMV and MCCB scores were higher in subjects with greater early-life urbanicity [20]. However, including the total GMV as a covariate may affect the benefits of early-life urbanicity. Thus, we used a less strict standard to determine whether there are some positive influences on GMV of early-life urbanicity. Additionally, previous studies have reported that subjects raised in a rural environment had a larger GMV of the medial prefrontal cortex (MPFC) [20] but did not explore associated environmental factors, such as parental education and presence of siblings. Hence, we analyzed the brain structural data in greater detail, explored potential associated environmental factors in this study.

Some brain regions could have important effect on cognition related to individual development and social function. The dorsolateral prefrontal cortex (DLPFC) is a key cortical region plays an important role in executive function, such as working memory, cognitive flexibility, planning, inhibition and abstract reasoning [21]. The temporal pole is another interested region, it plays an important role in episodic memory, semantic memory [22] and visual perception processing system [23, 24]. In our previous study, local spontaneous brain activity (regional homogeneity, ReHo) mediated the influence of urbanicity on the speed of processing [25]. However, we do not know whether GMV mediates the relationship between urbanicity and cognition. Hence, we investigated the effects of GMV changes affected by childhood urbanicity on cognition using mediation analyses in this study.

We have expanded earlier finding of an increased total GMV & cognition with more early-life urbanicity, and increased MPFC GMV with more early-life rural experience [20]. We found a positive relationship between early-life urbanicity and DLPFC and TP GMVs in this study, revealing some benefits of early-life urbanicity. Regarding specific environmental factors, we found that having siblings was associated with increased MPFC GMV and that higher maternal education was associated with higher total and DLPFC GMVs. In addition, the total, MPFC and TP GMVs exerted a positive mediating effect on the relationship between early-life urbanicity and cognitive performance in adulthood, specifically in terms of working memory, reasoning and problem solving (Fig. 3).

The MPFC GMV appears to increase with early-life rural experiences, especially the GMV of the medial regions (BA11 and BA8). The direction of the early-life upbringing effect is consistent with that of a similar study in Germany [5]. The MPFC is believed to play an important role in social cognition [35]. In our sample, rural subjects had more siblings, and the presence of siblings was significantly associated with the GMV of BA11. Siblings may provide more peer companionship and social interactions in childhood, which may result in lower neural sensitivity to social stress [36]. Our Previous study found that lower MPFC activity in response to social status threat was correlated with higher trait anxiety and depression in subjects with urban childhoods but not in those with rural childhoods [20]. Our current findings suggest that the presence of siblings may be a protective environmental factor for MPFC function. Decreased GMV of the MPFC was observed in subjects with prior experiences of adversity and stress, which is considered a risk factor for psychiatric disorders [37]. In our sample, there was no mediating effect of the altered GMV in the MPFC (rural > urban) on the relationship between childhood urbanicity and cognition scores. Our findings suggest that rural subjects have similar educational and vocational achievement not through enhanced cognitive performance but rather through increased social resilience, since the main function of the MPFC is social cognition and decision-making [37].

In addition to previous positive effects of early-life urbanicity on the total GMV and cognitive performance, we identified larger DLPFC and temporal pole GMVs with more early-life urbanicity. We found that maternal education positively affected the total and DLPFC GMVs, which is consistent with the importance of maternal education for children’s outcomes [38, 39]. Maternal education has been associated with socioeconomic status [40] and could affect the quality of cognitive stimulation in the home, such as mother–child interactions, the availability of books, computers, trips, parental communication and so on [41, 42]. The meta-analysis showed that activation of the lateral DLPFC is associated with working memory and episodic memory, which are two key features of cognition [43]. The mediating effect of the DLPFC GMV on the relationship of early-life urbanicity with the MCCB total score (including working memory, reasoning and problem solving) adds to our knowledge about the function of this brain region. Previous studies have shown that the functions of the DLPFC include branching and recollection of attention [44]. Our finding validates this hypothesis since the process of solving mazes required subjects to keep the goal in mind while exploring. In a study on patients with Alzheimer's disease, changes in the volume of the temporal pole were related to cognitive impairment [45, 46]. We found that the temporal pole GMV mediated the relationship between childhood urbanicity and reason and problem solving, which suggests that it is important for cognition.

The GMV and cognition differences between individuals with urban and rural childhoods may reflect nutritional and socioeconomic status differences. Although there was a fourfold improvement in per capita consumption from 1990 to 2009, the period in which most of our subjects were children or adolescents [47], the urban‒rural income gap increased simultaneously, with urban incomes rising markedly relative to rural incomes [48]. There are studies showing that family income and poverty status are powerful correlates of the cognitive development and behavior of children, even after accounting for other differences such as maternal education [49]. Studies have shown that approximately 39% of infants and toddlers (ages 0 to 3 years) born and raised in rural Chinese villages exhibit cognitive or psychomotor delays [50]. It seems that this pattern may be present even in those with educational achievements similar to those of their urban peers. However, these conclusions are speculative since we did not directly assess early-life stress and family income, although we inferred family income and early-life stress according to parental education and social development patterns. With the comprehensive poverty alleviation efforts implemented in China, this effect may weaken in the future.

This study has several limitations. First, the most rural group of subjects was slightly older and had more years of education than the other groups of subjects. Although we controlled for age and years of education during the analysis, there may have been residual influences on the results. Second, we recruited healthy individuals, and the influence of the early-life environment may differ in patients with mental disorders. Third, since we recruited relatively highly educated people currently living in Beijing, our sample is not representative of people with low education levels and living in rural areas. Forth, although we tried our best to include all objective factors we could obtain, the factors we included might not sufficient to cover all the aspects of the complex childhood environment compounds. Further research needs to be done in more diverse samples from clinical and nonclinical populations and on additional variables that could directly reflect early-life urbanicity.

In conclusion, we found a positive correlation of early-life urbanicity with cognitive scores and the total, DLPFC and TP GMVs and a negative correlation with the MPFC GMV. The increased total, DLPFC and TP GMVs partly mediated the relationship between early-life urbanicity and cognitive scores. In addition, the presence of siblings among individuals was associated with higher MPFC GMV, while maternal education was associated with better outcomes of total and DLPFC GMVs. On the one hand, this study replicates previous findings regarding the benefits of rural childhoods; on the other hand, this study suggests that there are also benefits from urban childhoods regarding brain and cognitive function.