Are Teachers Biased against Black Children? A Study of Race, Amygdala Volume, and Problem Behaviors

Introduction: While the amygdala has a core role in behaviors, less is known about racial variation in the association between amygdala volume and teachers’ behavioral rating of children. According to the Minorities’ Diminished Returns (MDRs) phenomenon, the effects of individual-level risk and protective factors tend to be weaker for Black than White children due to structural factors such as social stratification and racism. Purpose: Built on the MDRs framework and conceptualizing race as a social rather than a biological factor, this study explored racial variation in the magnitude of the effects of amygdala volume on teachers’ behavioral ratings of children. Methods: For this cross-sectional study, we used baseline socioeconomic data and structural magnetic resonance imaging (sMRI) data of 4305 American children ages 9-10 who had participated in the Adolescent Brain Cognitive Development (ABCD) study. The primary outcome was the teachers’ behavioral rating of the child. The independent variable was amygdala volume. Age, sex, parental education, parental marital status, and ethnicity were the covariates. Race was the moderator. We used mixed-effect models for data analysis to adjust for the participants’ nested nature within families and study sites. Results: Teachers rated children with larger amygdala volumes as having lower behavioral problems. The concordance between size of amygdala volume and teachers’ behavioral rating of the child was modified by race. For while children, teachers reported the children to have lower behavioral problems when they had a large amygdala. For Black children, teachers reported high behavioral problems across all amygdala sizes. Conclusions: The results can be explained in two ways. The first explanation is minorities’ diminished returns hypothesis (MDRs). In line with MDRs, due to structural inequalities and school segregation, a large amygdala www.scholink.org/ojs/index.php/jetss Journal of Education, Teaching and Social Studies Vol. 3, No. 1, 2021 2 Published by SCHOLINK INC. would result in a more favorable behavioral rating of the White children than Black children, as we expect an unequal effect of equal resources across racial groups in the presence of racism. The second explanation is systemic bias of teachers against Black children: meaning that due to their anti-Black bias, teachers report high behavioral problems in Black children, across all amygdala sizes (behavioral profiles). That means, race may trigger some cues and biases in the teachers, so they do not pay attention to the details of the behavioral profile of the Black child. For White children, however, in the absence of such racial bias, teachers behavioral rating of a child reflects the child’s amygdala size.


Participants and Sampling
Participants of the ABCD study were children who were between ages 9 and 11 years. Children in the ABCD study were recruited from multiple cities across multiple states. Overall, participants were enrolled from 21 sites. The primary source of recruitment for the ABCD sample was US school systems.
The sampling protocol of the ABCD study is described in detail here (Garavan et al., 2018). A total number of 4305 participants entered this analysis. Our analysis's eligibility included valid data on race, ethnicity, demographics, parental education, parental marital status, and children's amygdala volume.
Participants were included in this analysis regardless of their race or ethnicity.

Study Variables
Amygdala volume. The independent variable was the child's left amygdala volume (mm^3), measured by structural MRI at rest. More information on sMRI in the ABCD is available here (Hagler et al., 2019).
Our outcome had a normal distribution, as shown in Appendix 1. Appendix 2 shows the name of variables in the DEAP.

Moderator
Race. Race was self-identified by the parents. Race was a categorical variable: Black/African American, Asian, Mixed/Other, and White/Caucasian (reference category).
Age, sex, ethnicity, and parental marital status were the confounding variables. Parents reported the child's age and were calculated as months between the date of birth and the study's date. Sex of the child Linear regression in DEAP is based on mixed-effect models, given participants are tested to families, and families are nested to sites. The primary outcomes were internalizing, externalizing, and overall b behavioral problems (n=3). The independent variable as amygdala volume. Covariates were age, sex, parental marital status, parental education, and ethnicity. As such, in all our models, we controlled for the effects of families as sites. Our multilevel modeling approach is shown in Appendix 3. These models are run in a nested fashion, and small variations distinct them at each step. Model 1 tested the additive effects of amygdala size and race, with the same covariates, without interaction terms. Model 2 tested the interaction between amygdala size and race. We checked a wide range of assumptions, including normal distribution of our outcome, lack of collinearity between predictors, and the distribution of errors for our model and the association between observed and theoretical quantiles of our model (Appendix 4).

Ethical Aspect
Our secondary analysis was found by the Charles R Drew University of Medicine and Science (CDU) Institutional Review Board (IRB) to be exempt from a full IRB review. However, the original ABCD study underwent an Institutional Review Board (IRB) in several institutions, including but not limited to the University of California, San Diego (UCSD). The IRB in multiple institutions approved the study protocol, and all children provided assent and parents signed consent. 17% were other/mixed race. Most participants (84.4%) were non-Latino, and 15.6% were Latino. The mean left amygdala volume was 1567 mm 3 (SD = 233 mm 3 ). Table 1 also compared study variables by race. Racial groups did not differ in age or sex, but they differed in Hispanic ethnicity. Left amygdala size, internalizing behavioral problems, externalizing behavioral problems, and total behavioral problems varied across racial groups. All these problems were highest in Blacks and lowest in Whites and Asians. Left amygdala size was smallest in Black and largest in White children.

Externalizing Problems
As shown by Table 2, the amygdala size did not affect externalizing problem behaviors, when all confounders were controlled. As shown by

Discussion
Overall, we found inverse associations between amygdala size and internalizing and overall problem behaviors. However, the amygdala volume's effect was weaker for Black than White children. We found statistical interactions between amygdala size and race on internalizing, externalizing, and overall problems, reflecting Black children's disadvantage relative to White children in gaining behavioral outcomes from their amygdala size.
Our result (interaction) can be explained in two ways. First, we can explain it through the minorities' diminished returns (MDRs) lens. In line with MDRs, due to structural inequalities and school segregation, a large amygdala would result in more favorable behavioral profile of White than Black children. This is because of unequal effects of equal resources across racial groups in the presence of racism (Assari, 2017d; behavioral profile of the Black child (Assari & Caldwell, 2018c;Chavous, Rivas-Drake, Smalls, Griffin, & Cogburn, 2008). For White children, however, in the absence of such racial bias, teachers behavioral rating of a child reflects child's amygdala size.
We showed that while family SES is controlled, a larger amygdala size was associated with fewer problem behaviors. The effect of SES on brain development is well established (Farah, 2017). Many investigators such as Farah (2017), Noble, Norman and Farah (2005), and Lawson, Duda, Avants, Wu and Farah (2013) have shown an association between SES and various aspects of brain development.
For example, Lawson used structural MRI and SES data from a sample of 283 healthy children from the Study of Normal Brain Development and established a positive link between family SES and prefrontal cortical thickness in children. After confounders were controlled and multiple comparisons were adjusted in their study, parental education significantly predicted cortical thickness in the left superior frontal gyrus and right anterior cingulate gyrus. They argue that some cortical changes may mediate (explain) the effects of SES and cognitive function of healthy, typically developing children (Lawson et al., 2013). Sowell has also conducted multiple studies on the developmental change of cortex (Sowell, Thompson, & Toga, 2004;Sowell, Trauner, Gamst, & Jernigan, 2002). Her team has also studied cortical differences that correlate with demographic factors (Sowell et al., 2007), psychopathologies (Sowell et al., 2003) as well as environmental factors (Sowell et al., 2008). Most recently, Sowell used the ABCD data and among 9712 9 and 10-year-old children, showed a stronger negative association of living in high-lead-risk census tracts in children from lower SES families compared to higher SES families. Increased lead exposure was associated with smaller cortical surface area, smaller cortical volume, and lower cognitive test scores, but these associations were larger for needy families. Based on her findings, a change in the lead at the neighborhood level may generate a larger gain in low-SES brain development than high-SES children (Marshall et al., 2020).
Our second finding is that an increase in amygdala size may be associated with a smaller improvement in problem behaviors for Blacks than Whites is an extension of the MDRs literature. Past research using the ABCD data  both associated with a larger protective effect for depression , and anxiety (Assari, Caldwell, & Zimmerman, 2018), aggression , tobacco use , school attachment (Assari, 2019b), school performance ( Caldwell, 2020b), ADHD , impulsivity (Assari, Caldwell, & Mincy, 2018a), inhibitory control , stress , obesity (Assari, Thomas, et al., 2018), physical health , for White than Black children.
The pattern observed in this paper and the summary provided describe Blacks' Diminished Returns   immigrant Assari, 2020e), and even marginalized White families who were living in poor neighborhoods , this pattern is called Marginalization-related Diminished Returns (MDRs) .
Based on this model, any social identity, either visible (e.g., race) or invisible (LGBT status), would be associated with diminished SES returns for the participants.
To give examples of MDRs, we have found a higher prevalence of depression in high SES Black children and adults (Assari, 2017c;Assari, Gibbons et al., 2018a;Assari, Gibbons et al., 2018b;Assari, Lankarani, & Caldwell, 2018). In Flint Michigan, White children from married families have protected future symptoms of anxiety; however, this was not the case for Black children whose parental marital status does not protect against anxiety symptoms (Assari, Caldwell, & Zimmerman, 2018). In the Fragile Families and Child Wellbeing Study (FFCWS), which followed a sample of Black and White children from birth to age 15, we found that parental education and household income at birth better changed the mental and physical health outcomes of White than Black children when they were 15 years old. For example, while White children with high SES families were protected against ADHD, this protection was statistically smaller for Black families .
Alternatively, in the Population Assessment of Tobacco and Health (PATH) data, we have found that high SES Black children remain at risk of tobacco use. However, White children with high SES are protected against tobacco use . In the Monitoring the Future (MTF) data, we found the highest GPA in high SES White children. For Black children, however, GPA was staying low even when SES was high (Assari, Boyce, Bazargan, & Caldwell, 2020b). Finally, in the PATH data, we found that high SES White children are protected against aggression, obesity, tobacco use, and chronic disease. For Black children in the PATH data, family SES was not very protective .
The observed MDRs can be attributed to societal and structural factors such as racism. The MDRs framework argues that structural racism, social stratification, and contextual inequalities are the main reasons individual-and family-level resources and assets show diminishing returns for Black families compared to White families (Assari, 2017d). As long as higher-level barriers hinder Black families, discrimination is rampant (D. L. Hudson, Bullard, et al., 2012;, and Black families pay extra costs for their upward social mobility (D. al., 2020), individual-level resources and assets (either SES or brain structure) will continue to show smaller protective effects for Black children than White children. In this case, amygdala size shows weaker association with behavioral profile of Black children because Black children experience high levels of stress and discrimination , live in high risk neighborhoods , attend high risk schools , and spend time with high risk peers  and relatives . In such context, an individual level factor may operate as a weaker determinant Assari, 2017d), because every person is at risk, regardless of their resilience or risk factor status.

Limitations
A few limitations can be listed. This study only investigated correlates of amygdala volume, not other aspects of the amygdala. Amygdala diffusivity (dMRI), function (fMRI), and connectivity (fMRI) with other brain networks also have a role in the behavioral profile of youth. We also investigated the left amygdala without studying the asymmetry between the right and left amygdala.
Any study that investigates race and brain development has the potential to be mis-interpreted. As such, we distant ourselves from any racist interpretation of these results. We do not believe that racial groups are biologically different. We do not study brain structure of racial groups to suggest that one race is superior than the other race. We studied brain structure but we still believe that the observed differences are shaped by social forces that hinder Black communities. Studying mechanisms of social inequalities at the brain/biological level provides more convincing and stronger evidence regarding the deeply interwoven inequalities that are embedded in the US society.  , nativity , sexuality (Assari, 2019a), and even place .

Implications
Under racism, the effect of brain structures such as the amygdala are weaker in Black children than White generate outcomes for the historically oppressed group. As such, at the same level of resources, Black children do worse than White children. As a result, to promote brain health equity, eliminating the racial gap in resources may not be enough. It is unknown how residential segregation, labor market discrimination, risk in peers, and quality of schooling and neighborhoods explain why Black children remain at risk of low behavioral and emotional problems regardless of their amygdala. There is also a need to study structural and social factors that cause these MDRs, in addition to public and social policies to reduce such risk for children in middle-class Black families.

Conclusion
While larger amygdala size is associated with lower internalizing and overall behavioral problems for children reported by teachers, these effects are weaker for Black children than White children. We not only see MDRs for the effects of SES on the amygdala ; we also see MDRs for the effects of the amygdala on behaviors. Thus, MDRs can be seen for the effects of environmental inputs on brain, and also for the effects of brain on behaviors. Our approach to eliminating health and behavioral inequalities requires efforts beyond equalizing resources. As far as racism and segregation exist, resources and assets, including brain structures, show a weaker predictive role for Black families than White families. A true solution to health inequalities includes strategies to reduce racism in the US.

Conflicts of Interest:
The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
Acknowledgments: Data used in the preparation of this article were obtained from the Adolescent Brain Cognitive Development (ABCD) Study (https://abcdstudy.org), held in the NIMH Data Archive (NDA).
This is a multisite, longitudinal study designed to recruit more than 10,000 children aged 9-10 years and follow them over 10 years into early adulthood. The ABCD Study is supported by the National Institutes implemented the study and/or provided data but did not necessarily participate in analysis or writing of this report. This manuscript reflects the views of the authors and may not reflect the opinions or views of the NIH or ABCD consortium investigators. The ABCD data repository grows and changes over time.