Involvement in Sports, Hippocampal Volume, and Depressive Symptoms in Children

Abstract

Background:

Recent studies have found that higher levels of exercise are associated with fewer symptoms of depression among young people. In addition, research suggests that exercise may modify hippocampal volume, a brain region that has been found to show reduced volume in depression. However, it is not clear whether this relationship emerges as early as preadolescence.

Methods:

We examined data from a nation-wide sample of 4191 children ages 9–11 years from the Adolescent Brain and Cognitive Development Study. The parents of the children completed the Child Behavior Checklist, providing data about the child’s depressive symptoms, and the Sports and Activities Questionnaire, which provided data about the child’s participation in 23 sports. Children also took part in a structural MRI scan, providing us with measures of bilateral hippocampal volume.

Results:

Sports involvement interacted with sex to predict depressive symptoms, with a negative relationship in boys only (t= −5.257, β= −0.115, p< 0.001). Sports involvement was positively correlated with hippocampal volume in both boys and girls (t= 2.810, β= 0.035, p= 0.007). Hippocampal volume also interacted with sex to predict depressive symptoms, with a negative relationship in boys (t= −2.562, β= −0.070, p= 0.010), and served as a partial mediator for the relationship between involvement in sports and depressive symptoms in boys.

Conclusions:

These findings help illuminate a potential neural mechanism for the impact of exercise on the developing brain and the differential effects in boys versus girls mirror findings in the animal literature. More research is needed to understand the causal relationships between these constructs.

Depression is associated with symptoms such as anhedonia, feelings of hopelessness and suicidal ideation and can be debilitating in people of all ages (1). Unipolar depressive disorders are the third leading cause of the global burden of disease (2). Much of the research on depression has focused on adults. However, 3.1 million adolescents in the United States had at least one major depressive episode in 2016 alone (1). As such, it is imperative that more research be focused on understanding the development of depression in children and adolescents. Work that can identify early risk factors or pathways for prevention or early intervention in children and adolescents is particularly crucial. The goal of the current study is to examine the relationship between one potentially relevant factor, exercise, and depression in children, as well as to examine the neural mechanisms that might mediate such a relationship.

Exercise has been shown to have a positive relationship to mental health. For example, active lifestyles have been positively associated with reduced symptoms of depression and anxiety, improved self-concept, and more effective coping with stress (3). Additionally, a randomized controlled trial using older adults found that after 20 weeks, clinical depression had resolved in 73% of adults assigned to an exerciser condition but in only 36% of adults assigned to a control condition (4). Similarly, a meta-analysis found that resistance training significantly reduced depressive symptoms among adults (5). Fewer studies have examined the relationship between involvement in exercise and mental health in children. However, a review of reviews found that higher levels of exercise were significantly associated with fewer depression symptoms among young people aged 18 and younger (6). Additionally, another study found that sports participation in high school was predictive of better school engagement, school performance, and self-esteem (7).

There are multiple aspects of exercise that may have beneficial impacts on children and their mental health. One reason that exercise has a positive relationship to mental health may be because it often occurs through engagement in sports. Engagement in sports may be because of the social support that can arise from being a part of a team. One study found that being involved in at least one activity (sports, art, music, or other activities) was associated with higher life satisfaction, better self-rated health, and a lower frequency of feeling “low” (8). Interestingly, the strongest associations with healthy development indicators were in adolescents engaged solely in sports and not other activities (8). This may suggest that there is something integral about sports that is causing their benefit over and above participation in organized activities in general.

Another mechanism by which exercise may relate to positive mental health is through its impact on the developing brain. Studies of rodents have found that aerobic exercise in the form of wheel running leads to increased neurogenesis and long-term potentiation in the dentate gyrus of the hippocampus (9). Furthermore, exercise promotes brain vascularization, changes in neuronal structure, neuronal resistance to injury, and increased levels of brain derived neurotropic factor (BDNF) in the hippocampus (10). Studies of humans show similar protective effects. In particular, aerobic exercise is associated with larger left and right hippocampi in elderly populations (11). In addition, a study of a 6-month aerobic exercise intervention in adults aged 60–79 years demonstrated increased regional gray matter volume in the frontal, parietal, and temporal lobes (12). Exercise also appears to have a beneficial effect on the brains of children. Greater aerobic fitness, measured through VO2 max, is related to increased left middle prefrontal cortex volume, and left precuneus and right occipital surface areas in male adolescents (13). Additionally, one study found that children who were more fit had greater bilateral hippocampal volumes, and that hippocampal volume mediated the relationship between fitness level and score on a relational memory task (14). Thus, it is important to understand the relationship between exercise and the hippocampus in preadolescent children.

The hippocampus has also been implicated in the development of depression (15–18). Hippocampal volume has been associated with depression symptom severity and illness duration (19) and the number of depressive episodes a person has experienced (20). Studies using younger populations have found similar associations between depression and hippocampal volume (21), with early onset adolescent depression associated with a 17% reduction in left hippocampal volume (22). In addition, a meta-analysis found that patients with childhood onset MDD had average reductions in hippocampal volume by 5.3% in the left hemisphere and 5.2% in the right hemisphere (23). The hippocampus is critical for the inhibition of the hypothalamus-pituitary-adrenal (HPA) axis via the presence of glucocorticoid receptors that are part of a negative feedback loop (24). When cortisol levels are high, excitotoxic effects in the hippocampus may contribute to problems inhibiting the HPA axis, resulting in even more cortisol and a problematic cascade (25). Depression has also been associated with the experience of life stress and altered stress reactivity, which may be at least part of the reason hippocampal volume may be reduced in patients with MDD (26, 27).

As reviewed, exercise may be positively related to mental health in children for a number of reasons, including the potential positive impact of social support through engagement in team sports and/or a positive impact on brain development. Here we focus on hippocampal volume given the combined animal and human literature that converges on the relationship between exercise and hippocampal structure and because of the putative association between hippocampal integrity and depression. To examine these hypotheses, we used data from 4191 nine-to-eleven-year-old children from across the United States participating in the Adolescent Brain and Cognitive Development study (ABCD). We predicted that a) greater involvement in sports, but not non-sport activities, would predict both fewer depressive symptoms and larger hippocampal volumes; b) the relationship between sports, depression, and hippocampal volume would remain even if we controlled for involvement in non-sports related activities; c) if at least part of this effect was due to the social support engendered by team sports, this effect would be largest in the case of team sports over and above individual sports; d) children with depressive symptoms would have smaller hippocampal volumes than their healthy peers; and e) hippocampal volume would serve as the mediating variable between sports participation and depressive symptoms.

Method

Participants

Participants took part in the ABCD study, a longitudinal study tracking over 11,800+ children ages 9–11 years old from 21 different sites across the United States. The sampling strategy to approximate national norms is described elsewhere (28). All study procedures were approved either by the centralized IRB at the University of California San Diego or by an individual participating site’s IRB. All parents signed informed consent and all children provided written assent prior to participation in the study. While there were 4524 children originally in the data set, 333 had to be removed for the current analyses because the brain data was unusable (N=328) or they did not fall into the male or female binary (N=5). Of those that did not define their gender as male or female, one person was trans female, one person described their gender as “different,” and 3 people did not report gender. Therefore, 4191 children’s data was used in the analyses reported below. For a comparison of demographics of male and female participants, please see Table 1. For information regarding demographics of the included versus excluded participants, please see Supplemental Table S1.

Table 1.

Demographics as a Function of Sex

Characteristics	Male (N=2197)	Female (N=1994)	Statistic	P Value
	Mean ± SD [Min-Max] or %	Mean ± SD [Min-Max] or %	T Score
Age (In Months)	120.44 ± 7.350 [108–132]	120.27 ± 7.225 [108–132]	0.740	0.459
Number of Total Activities	3.70 ± 2.682 [0–18]	3.93 ± 2.839 [0–20]	−2.706	0.007
Number of Sports	2.76 ± 2.056 [0–14]	2.75 ± 2.134 [0–18]	0.065	0.948
Number of Non-Sports	0.94 ± 1.137 [0–6]	1.17 ± 1.212 [0–6]	−6.456	<.001
CBCL Depression T Score	53.79 ± 5.954 [50–89]	52.98 ± 5.124 [50–86]	4.745	<.001
Average Hippocampal Volume (in cubic mm)	4208.69 ± 395.06 [2779.45–5675.40]	3938.24 ± 378.0 [2111.30–5486.60]	22.034	<.001
Family Income Level			0.005	0.996
Under $25,000	11	9.9
$25,000 to $49,999	12.1	13.1
$50,000 to $99,999	27.8	27.9
$100,000+	41.2	41.6
Parent Education			0.394	0.693
High School/GED or less	13.9	13.5
Some College or Associates Degree	29.3	26.5
Bachelor’s Degree	30.0	31.7
Greater than Bachelor’s	26.5	28.5
			X2 Value
Team Sport (Broad Definition)	84.0	81.9	3.36	0.070
Team Sport (Restrictive Definition)	74.9	52.4	231.50	<.001
Individual Sport	61.6	55.1	18.29	<.001
Structured Sport	85.2	83.6	2.05	0.159
Race
Caucasian	81.7	80.4	1.17	0.286
African-American	13.9	15.9	3.52	0.062
Other	4.4	3.7
Hispanic	20.4	19.5	0.53	0.484

Measures

The ABCD study included a number of measures of mental health related constructs (29). The Child Behavior Check List (CBCL), completed by the child’s caregiver, provides information about mental health problems in the child. We examined the child’s Depressive age corrected T Score. For analyses examining anxiety, we used the child’s Anxiety age corrected T Score. The CBCL test-retest reliability over 8 days was r= .84 for the Depressive score and r=.80 for the Anxiety score (30).

The Sports and Activities Involvement Questionnaire measures lifetime involvement in sports, activities such as music and dance, and other hobbies (see Supplemental Table S2). It provides a parent report of the frequency, duration, and type of activity, in addition to whether it is organized, private, individual, or structured. For the purposes of this study, we examined the child’s involvement in 23 different sports (see Supplemental Table S2 for list). We further divided the data into the categories of team sport (broad definition), team sport (restrictive definition), individual sport, and structured sport. In the team sport (broad definition) category, the child engaged in the sport at school or in an organized outside league. In the team sport (restrictive definition), the child engaged in the sport at school or in an organized outside league, but it needed to be one of the following sports where the sport was played as a team activity: baseball, basketball, field hockey, football, ice hockey, lacrosse, rugby, soccer, or volleyball. In the individual sport category, the child engaged in the sport on their own time or through private lessons. Finally, in the structured sport category, the child engaged in the sport at school, through an organized outside league, or through private lessons. These categories of sports were not mutually exclusive. For example, participation in baseball would count as both a team sport (broad definition) and a team sport (restrictive definition). The non-sport activities examined are also listed in Supplemental Table S2.

MRI Data

Brain data were collected on 3T scanners across 21 sites, including Siemen’s PRISMAs, GE 750s and Phillips Achieva. The parameters for the T1 and T2 acquisitions for each of these platforms are listed in Table 2 of the ABCD imaging paper (31). The T1 and T2 images were corrected for gradient nonlinearity distortions, and the T2 images were registered to the T1 images using mutual information and atlas based registration (32–34). Hippocampal and whole brain volumes were computed from the preprocessed T1 images using FreeSurfer version 5.3.0. Hippocampal and whole brain volumes came from the “aseg” atlas. As we did not have hypotheses about left versus right hippocampus, we combined them to form an average hippocampal volume variable, though analyses of left and right hippocampus separately provided similar results and are included in Supplemental Table S7. The quality control of the T1 and T2 data are described in the ABCD image processing paper (35).

Table 2.

Relationship Between Involvement in Sports and Depressive Symptoms

IV	Overall		Sex Interaction	Male		Female
	β	T Score	FDR P Value	β	T Score	β	T Score
# of Activities	−0.072	−4.481***	0.023*	−0.097	−4.391***	−0.021	−0.850
# of Sports	−0.089	−5.590***	0.023*	−0.115	−5.257***	−0.041	−1.655
# of Non-Sports	−0.007	−0.428	0.350	--	--	--	--
Team Sport (Broad)	−0.093	−5.816***	0.0035**	−0.129	−5.888***	−0.031	−1.248
Team Sport (Restrictive)	−0.112	−6.909***	0.0035**	−0.139	−6.401***	−0.059	−2.540
Individual Sport	−0.076	−4.897***	0.023*	−0.103	−4.789***	−0.037	−1.601
Structured Sport	−0.097	−6.092***	0.0047**	−0.133	−6.070***	−0.038	−1.525

Note. All analyses are using covariates of age (in months), race, ethnicity, maternal education, and family income.

^*p < .05.

^**p < .01.

^***p < .001

Analyses

The analyses used linear regressions that allowed us to provide standardized beta weights as a measure of effect size. However, the sample included a number of sets of twins. Thus, to account for this in our analysis, we confirmed analyses using general linear models with participants nested within sites and families. We carried out analyses for each sports category predicting depression and hippocampal volume, and for the relationships between hippocampal volume and depression. We covaried for race, ethnicity, age, parental education, family income, and intracranial volume and included both sex and the interactions between sex Significant interactions with sex were and independent variable of interest as predictors. followed up with regressions within each sex.

We additionally carried out analyses examining the relationship between non-sports and depression and between sports and depression while controlling for involvement in non-sport activities. Likewise, we carried out analyses examining the relationship between non-sport activities and hippocampal volume. Finally, we examined the relationship between involvement in team sports and hippocampal volume, while controlling for individual sports. To determine if our results were specific to depressive symptoms, we additionally examined the relationship between involvement in sports and activities and anxiety t score while controlling for depressive symptoms and vice versa. In all analyses, we used the False Discovery Rate (FDR) to correct for multiple comparisons (36, 37). The mediation analysis used the process macro in SPSS. See Supplemental Materials for additional analyses of sports “dose” relationships.

Results

Table 1 describes the demographic characteristics of the children in the sample that we analyzed. 3.8% of children scored in the borderline clinical range for depression and 2.8% scored in the clinical range. As shown in Table 2, greater participation in every category of sport, but not non-sports activities, was associated with fewer depressive symptoms, even when correcting for socioeconomic status, maternal education, race, ethnicity, and age, with all relationships passing FDR correction. However, we also saw significant interactions with sex for all of the sports/activities predictors, with all but non-sports activities passing FDR correction. Follow up analyses (Table 2) within sex indicated significant relationships between every type of activity/sport and depression in boys, but after FDR correction, no significant relationships in girls (Figure 1 and Supplemental Figure S1). Rerunning the analyses controlling for involvement in non-sport activities did not impact any of the results. Moreover, as shown in Supplemental Table S3, no subcategory of non-sport activities (group, individual, or structured) had a significant relationship to depressive symptoms after FDR correction.

Figure 1.

A: Graph illustrating the relationship between depression symptom severity and the number of sports in which a child is involved. Regression lines are shown separately for girls and boys. Each dot represents more than one child. B: Bar graph illustrating the relationship between participation in team sports – restrictive definition and depression symptom severity, plotted separately for males and females.

Interestingly, the relationship between sports involvement and mental health appeared to be specific to depression and not anxiety. When we tested the association between anxiety and each of the sports and activities categories (Supplemental Table S4) using depression as a covariate, the relationship was not significant for any activity. Further, as shown in Supplemental Table S5, when we tested the relationships between activities and sports to depression using anxiety as a covariate, all results remained significant.

We next examined the relationships between the different types of sports and activities and hippocampal volume. Greater participation in every category of sport/activity, except for individual sports and non-sports activities, was associated with larger hippocampal volume (Table 3). Involvement in team sports (both broad and restrictive) was associated with greater hippocampal volume even when we controlled for involvement in individual sports. Interestingly, unlike depressive symptoms, there were no significant interactions with sex for any of the activity/sports variables. Follow up analyses split by sex indicated that all of the relationships were significant for boys, with a similar trend for girls (Figure 2). Rerunning the analyses controlling for involvement in non-sport activities did not change the results. As shown in Supplemental Table S6, no subcategory of non-sport activities had a significant effect on average hippocampal volume. Finally, we tested the associations between involvement in sports and activities and left and right hippocampal volume separately. The relationships were similar in the left and right hippocampus (Supplemental Table S7), with the only difference being the relationships between number of activities and number of sports on right hippocampal volume, which were trend level.

Table 3.

Relationship Between Involvement in Sports and Average Hippocampal Volume

IV	Overall (Male and Female)			Sex Interaction FDR P Value
	β	T Score	FDR P Value
# of Activities	0.035	2.806	0.007**	0.803
# of Sports	0.035	2.810	0.007**	0.803
# of Non-sports	0.019	1.517	0.129	0.857
Participation in a team sport (broad)	0.042	3.429	0.002**	0.803
Participation in a team sport (restrictive)	0.042	3.354	0.002**	0.803
Participation in an individual sport	0.020	1.630	0.120	0.803
Participation in a structured sport	0.044	3.567	<.001***	0.803

Note. All analyses are using covariates of age (in months), race, ethnicity, maternal education, family income, and intracranial volume.

^*p< .05.

^**p < 0.01.

^***p < 0.001

Figure 2.

A: Graph illustrating the relationship between hippocampal volume and the number of sports in which a child is involved. Regression lines are shown separately for girls and boys. Each dot represents more than one child. B: Bar graph illustrating the relationship between participation in team sports-restrictive definition and hippocampal volume, plotted separately for males and females.

The analyses predicting depressive symptoms from average hippocampal volume (see Figure 3) showed both a significant main effect (t=−2.428, β=−0.052, p=0.015) and a significant interaction with sex (t=2.101, β=0.355, p=0.036). Follow-up analyses within sex indicated that hippocampal volume was negatively associated with depressive symptoms in males (t=−2.562, β=−0.07, p=0.010) but not in females (t=−0.830, β=−0.025, p=0.41). Furthermore, depressive symptoms in males were predicted separately by both left hippocampal volume (t=−2.572, β=−0.067, p= 0.01) and right hippocampal volume (t=−2.122, β=−0.058, p=0.034).

Figure 3:

Graph illustrating the relationship between depression symptom severity and hippocampal volume. Regression lines are shown separately for girls and boys. Each dot represents more than one child.

Finally, we used a mediation analysis to further test hypotheses about the relationship between sports and activities, hippocampal volume, and depressive symptoms in boys. As shown in Table 4, hippocampal volume partially mediated the relationship between depression and the number of activities, the number of sports, both definitions of team sports, and structured sports, though these predicted relationships would not survive FDR correction.

Table 4.

Mediation Analysis of Hippocampal Volume with Sports and Depression in Boys

IV	Indirect Effect	Lower Confidence Interval	Upper Confidence Interval
# of Activities	−0.0065	−0.0153	−0.0002
# of Sports	−0.0082	−0.0199	−0.0004
Team Sport (Broad)	−0.0545	−0.1245	−0.0048
Team Sport (Restrictive)	−0.0391	−0.0946	−0.0026
Structured Sport	−0.0597	−0.1367	−0.0033

Note. All analyses are using covariates of age (in months), race, ethnicity, maternal education, family income, and intracranial volume.

Discussion

The goal of the current study was to better understand the relationship between sports and activities involvement to preadolescent children’s brain development and mental health. Understanding the potential mechanisms of the relationship between sports and depression is critical for developing informed public health guidelines about the importance of sources of exercise at a young age. We hypothesized that involvement in sports, but not non-sport activities, would predict fewer depressive symptoms and this proved to be true across all subcategories of sports. However, surprisingly, these results only held in boys, but not girls. Moreover, the beneficial effect of sports was specific to depression and not anxiety symptoms. Additionally, we hypothesized that involvement in sports would be associated with larger hippocampal volume, and we found evidence for this relationship for all types of sports except individual sports in both boys and girls. However, hippocampal volume was associated with depressive symptoms in males only, and partially mediated the relationships between involvement in multiple types of sports and depressive symptoms. Each of these results will be discussed in more detail below.

Our finding that greater sports involvement, but not non-sport activity involvement, was associated with less depression in boys suggests that exercise, a factor inherent to sports involvement, could be having anti-depressant effects. This would be consistent with studies that found positive associations between active lifestyles and reduced depression, in addition to studies demonstrating exercise as an effective antidepressant (3–6). However, these studies did not find sex differences. Our findings of sex specificity may relate to the age of our sample. Puberty tends to have an earlier onset in girls than boys. Thus, during preadolescence, many boys may not yet be going through puberty, but girls may be starting to experience hormonal changes that may be having an impact on depression and might be obscuring the relationship between sports, hippocampal volume, and depression. Another contributor to this sex difference could be that young girls may be subject to very different cultural attitudes about sports than boys. If a child is engaging in a sport for fun, and it is not a stressful activity, it will have a very different effect than if it is a hyper-competitive environment. Moreover, if girls are engaging in sports in order to lose weight and fit society’s thinness standards, the pressure may no longer make the sport a fun experience, taking away the antidepressant effects.

We also found that involvement in all types of sports except for individual sports and non-sport activities was related to hippocampal volume in both males and females. As involvement in sports usually implies exercise, this finding is consistent with numerous studies linking exercise to greater hippocampal volumes in both humans and animals (9–12, 14). However, the fact that involvement in individual sports did not predict hippocampal volume suggests that there may be a beneficial component of being on a team or being part of a structured program over and above the benefit from exercise for children. This idea is further supported by the fact that when we examined the relationship between team sports (both broad and restrictive definitions) and hippocampal volume controlling for involvement in individual sports, involvement in team sports was still a significant predictor of hippocampal volume. Research specifically looking at the effects of team sports over solitary exercise on hippocampal volume will be necessary to fully understand how these relationships manifest in children.

Finally, we found that hippocampal volume was associated with depression in boys only, which was partially consistent with past research. Numerous studies have demonstrated greater depressive symptoms as being associated with reduced hippocampal volume (15–17). Hippocampal volume has also been shown to be associated with depressive symptom severity and illness duration (19) and number of depressive episodes (20). However, to our knowledge, this sex difference has not been previously reported in a preadolescent sample. On the other hand, at least two studies have found that depressed males showed smaller left hippocampal volume than male controls, but this relationship did not hold in females (38, 39). While most studies of adolescents did not report sex differences (21), one study did find a stronger reduction in hippocampal volume in depressed males, but the reduction was still present in females (22). One potential reason for the differences in findings may be sample size. Most studies of hippocampal volume in depressed adolescents, with sample sizes of less than 100, are not very well powered to detect sex differences (21, 22). However, the ABCD data set, with over 4,000 children, has the statistical power to detect sex differences that a smaller data set might not allow.

Our finding that hippocampal volume was more strongly related to depression in boys than girls is at least indirectly consistent with some of the animal literature related to depression. Animal models often use interventions like maternal deprivation or limited bedding and nesting to induce depressive-like states in the rodents. A number of these studies have found poorer behavioral and structural/neurological outcomes in male rodents, with female rodents relatively spared (40–42). This sex difference may be related to hormone levels, as studies in both rats and humans have demonstrated neuroprotective effects of estrogen (43–45). Moreover, testosterone has been shown to increase vulnerability to neurotoxic processes in rats (46). Thus, more research must be done in humans to better understand how the brain develops differently in males and females, how environment and other activities might differentially influence brain development across sexes, and how this relates to mental health.

Limitations

The way in which the data was collected allowed us to determine the number of sports in which a child had been involved in across their lifetime, but it was difficult to examine how many activities were occurring at the same time. (See the Supplement for analyses of “dose” relationships during the most intense period of sports engagement.) In addition, it would have been useful to have other measures of physical fitness, such as VO2 max. We did not ask the children about the level of competitiveness of their sport, and this could have influenced the degree to which they experienced positive support from the activity. In addition, the current analyses were based on cross sectional data, and thus cannot address causality. However, the ABCD study is longitudinal, and it will be important to examine how these relationships change over time, as this will give us further information about the potential causal relationships between involvement in sports, hippocampal development, and mental health outcomes. Lastly, while we found significant relationships that would be predicted by the prior work in adults and animals, the magnitude of the effect sizes were small, with significance achieved in part through the very large sample sizes. Nonetheless, the presence of these predicted relationships in a large population sample is intriguing and provides motivation for future studies that can better address questions of causality.

Future Directions

All of our results are associations, meaning that we cannot infer causality. However, the ABCD study is longitudinal, and future work can utilize this longitudinal data to tease out causality in the relationships between involvement in sports, hippocampal volume, and depressive symptoms. For example, it is unclear if smaller hippocampal volume is a risk factor for depression or a consequence of onset of the disorder. A study of patients with varying numbers of past episodes of depression suggests that hippocampal volume decreases as a result of illness onset, but future analyses with the ABCD dataset will help determine the direction of this relationship in children (47). Furthermore, while involvement in sports and/or social activities may help reduce depressive symptoms, it is also possible that being depressed makes someone less likely to participate in activities, and this could further lead to hippocampal changes. Monitoring these relationships as the ABCD children progress through adolescence will be critical for determining the temporal dimension of these relationships. The majority of the children in this baseline ABCD sample were pre- or early puberty. It is possible that the relationships examined here will change as a function of puberty status, and in future analysis, it will be important to examine puberty levels in order to explore our sex difference finding.

Conclusion

This study was one of the first, to our knowledge, to show that involvement in sports is associated with the mental health and brain development of children as young as 9–11 years old. Involvement in sports was associated with fewer depressive symptoms in boys, and this relationship was partially mediated by hippocampal volume. These associations highlight potential causal mechanisms which can be investigated using longitudinal data. This is critical for public policy, because if involvement in sports does change children’s brain development through the impact on the hippocampus and related networks, there is a strong incentive to encourage children as young as 9–11 to participate in sports regularly.