Abstract
Background:
Substantial heterogeneity exists in how rearing environments influence youths’ socio-emotional outcomes. This heterogeneity, as suggested by the biological sensitivity to context (BSCT) and the differential susceptibility (DST) theories, is associated with emotional reactivity patterns and underlying neural functions. The present study investigated amygdalar reactivity to emotional stimuli as a neural signature that amplified the influence of rearing environments on youths’ socio-emotional outcomes.
Methods:
To increase replicability and generalizability, this investigation included two independent studies that methodologically complemented each other. Study I employed a large, national, and longitudinal dataset (the ABCD study; N=11,875). Study II used a community sample of youths (N=123) with multi-method and multi-reporter assessments.
Results:
In Study I, high left amygdalar reactivity to positive stimuli significantly amplified the impact of parental warmth on youths’ prosocial behaviors. In Study II, left and right amygdala reactivity to positive stimuli significantly intensified the associations between family functioning and youths’ internalizing problems. These findings were consistent with the BSCT/DST hypothesis because significant socio-emotional differences were observed at both negative and positive extremes of rearing environments. Additionally, Study II partially supported the diathesis-stress hypothesis by showing significant differences in youths’ vulnerability to negative family environments. Specifically, left amygdalar response to negative stimuli exacerbated the associations between unbalanced family functioning and heightened internalizing/externalizing symptoms. Left amygdalar reactivity to positive stimuli intensified the link between unbalanced family functioning and elevated externalizing problems.
Conclusions:
Among youths and adolescents, amygdalar emotional reactivity may serve as a biomarker of differential sensitivity to rearing environments.
Keywords: Amygdala, Early Life Stress, Parenting, Family Rearing Environments, Differential Susceptibility, Internalizing and Externalizing Symptoms
Introduction
Family rearing experiences play a critical role in youths’ socio-emotional development (e.g., 1, 2). However, substantial individual differences exist in how adverse or positive environments influence youths’ developmental outcomes (3, 4). According to the biological sensitivity to context theory (BSCT; 5, 6) and the differential susceptibility theory (DST; 7), youths’ differential sensitivity to environmental influences is partly attributed to the heterogeneity in emotional reactivity and underlying neurobiological processes (8). The present study examined amygdalar reactivity to emotional stimuli as a neurobiological moderator of the influences of rearing environments on youths’ socio-emotional outcomes.
Family Environments and Socio-Emotional Adjustment
Family environment is a multi-dimensional social context that significantly affects youths’ socio-emotional development (9). Unsupportive emotional climates induced by family conflict disrupt youths’ emotional security and foster the development of psychopathology (10, 11). Conversely, parental warmth fulfills youths’ affective needs and promotes prosocial behaviors (12). Based on the circumplex model of marital and family systems (13), family functioning (i.e., the quality of interactions among family members; 14) captures the full continuum (from negative to positive) and dimensionality (e.g., cohesion and flexibility) of rearing environments (15). Unhealthy/unbalanced family functioning, characterized by extremely low (i.e., disengaged or rigid) or high (i.e., enmeshed or chaotic) levels of cohesion and flexibility, is associated with increased risk for psychopathology (16). In contrast, healthy family functioning where cohesion and flexibility dimensions are balanced (not extreme) contributes to positive socio-emotional adjustment (17).
Despite the consistent links between caregiving experiences and socio-emotional outcomes, youths respond differently to similar family environments (3, 4). Some youths are highly sensitive to environmental inputs (18), while others are less responsive to family influences (5). This heterogeneity has been linked to youths’ differences in emotional reactivity (19, 20) and relevant neurobiological processes. In particular, the BSCT (5, 6) and DST (7) converge to suggest that a specific neural signature underlying emotion processing could interact with positive or negative rearing environments and induce qualitatively different behavioral adjustments, which follows a “for better or worse” pattern (5–7). Neural functions underlying heightened emotional reactivity may amplify youths’ benefits from positive experiences and promote optimal socio-emotional adjustment (i.e., “for better”; vantage sensitivity; 21). In contrast, these neural processes can exacerbate youths’ vulnerability to adverse caregiving environments and lead to elevated risks for psychopathology (i.e., “for worse”; diathesis-stress; 22).
Amygdalar Reactivity and Differential Sensitivity
Recent studies that examine the neural functions underlying differential sensitivity to environmental inputs are emerging (8, 23–25). One neural structure that may potentiate youths’ differential sensitivity is the amygdala, a critical brain region for emotion processing and stress response modulation (26). Elevated amygdalar reactivity has been long documented in response to both positive and negative emotional faces (26, 27). Individual’s perceptions of aversive and positive social stimuli are associated with emotional salience in the amygdala (27) and are further modulated through amygdalar feedback loops involving attention and vigilance (28). The amygdala also affects stress responses by directly modulating individual’s physiological stress responses (29).
Recent empirical studies have demonstrated that amygdalar responses to emotional stimuli is an indicator of young adults’ differential sensitivity to positive and negative environmental influences (30, 31). For example, Swartz and colleagues (31) reported associations between amygdalar reactivity to threatening facial stimuli and vulnerability to stressful life events in a sample of 340 college students. In another study with 310 low-income young men, Gard and colleagues (30) examined amygdalar reactivity to facial expressions as a sensitivity biomarker to the effects of socioeconomic resources. Employing rigorous quantitative tests (32, 33), findings from this study indicated that amygdalar reactivity significantly amplified the effects of socioeconomic resources on young adults’ antisocial behaviors and income attainment, which supported the “for better and for worse” hypothesis of BSCT/DST. Together, these two studies suggest that amygdalar reactivity to emotional stimuli may serve as a differential sensitivity biomarker among young adults.
Gaps in the literature still remains as previous empirical studies often focus on young adult populations and rely on modest samples with limited measurements of social environments. A rigorous examination of the BSCT/DST requires large samples because it involves probing the interactions between multi-dimensional environments and neurobiological markers using critical quantitative tests (32, 33). It is also critical to assess both positive and negative developmental outcomes and to employ measurements that capture the full spectrum of environmental inputs (5–7). No studies have yet tested the associations between amygdalar reactivity and differential sensitivity to family environments in large samples of pre-adolescents and adolescents.
The Present Study
Built on previous theoretical (5–7) and empirical (30, 31) work, the current study investigated the moderating effects of amygdalar response to emotional stimuli on the associations between family environments (positive & negative) and youths’ socio-emotional outcomes (positive & negative). First, we hypothesized that elevated amygdalar response to negative stimuli would amplify the associations between negative environments (i.e., family conflict & unbalanced family functioning) and maladaptive outcomes (i.e., increased psychopathology symptoms & reduced optimal outcomes). We then hypothesized that greater amygdalar response to positive emotional stimuli would intensify the relations between positive experiences (i.e., parental warmth & balanced family functioning) and adaptive outcomes (i.e., reduced psychopathology symptoms & increased positive outcomes). Lastly, we hypothesized that family environments would have non-significant influences on socio-emotional outcomes among youths with low amygdalar reactivity. Moreover, mixed findings have been reported on the lateralization (i.e., the tendency for certain brain functions to be specialized to one brain hemisphere) on amygdalar reactivity (34–36). Thus, we tested left and right amygdalar reactivity separately to explore laterality differences.
Hypotheses were tested using two studies that methodologically complemented each other. Study I used a national dataset of the Adolescent Brain Cognitive Development (ABCD) study, and Study II comprised data from rural Georgia families. While data from Study II were cross-sectional and of moderate sample size, Study I dataset was longitudinal and large, granting substantial statistical power and generalizability of findings. Regarding measurements, Study I used shortened surveys to measure family conflict and parental warmth, which, despite assessing both negative and positive components, did not capture the full spectrum of family experiences. In contrast, Study II employed a rigorous measurement to capture the full continuum and dimensionality (i.e., cohesion & flexibility) of family functioning. Using these two studies provided a comprehensive investigation of our hypotheses and increased replicability and generalizability of study findings.
Methods and Materials
Participants
The ABCD study included baseline (N = 11,875) and 1-year follow-up data from a sample of youths and primary caregivers (Release 3.0; 37, 38). Participants were recruited at 21 sites using probability sampling (37). The sampling and recruitment methods ensured a locally (i.e., sites) representative sample of youths aged 9–10, which combined and provided a close approximation to the national sociodemographic of this population (37). Case-specific weights were obtained using propensity-based methods (39) to compensate for existing selection biases. Study II used a sample of adolescents (12–14) and primary caregivers who resided in rural Georgia (N = 123). Table 1 presents participants’ demographics.
Table 1.
Demographics of Study Participants
Demographic Variables | Study I (N = 11,875) | Study II (N = 123) | |
---|---|---|---|
Youth | Biological Sex | 47.9% female | 54.5% female |
Age | 9 – 10, M = 9.48, SD = .51 | 12 – 14, M = 12.90, SD = .81 | |
Race/ethnicity | |||
White | 52.3% | 77.9% | |
Black | 15.1% | 12.2% | |
Hispanic/Latino(a) | 20.4% | 4.9% | |
Other | 12.3% | 5.7% | |
Involvement in Study | |||
One child in each family | 68.6% | 100% | |
Siblings | 13.4% | 0.0% | |
Twins | 17.7% | 0.0% | |
Triplets | 0.3% | 0.0% | |
Caregivers | Biological Sex | 89.0% female | 93.4% female |
Age | M = 39.97, SD = 6.84 | M = 43.25, SD = 6.41 | |
Education | |||
Below High School Diploma | 5.0% | 3.3% | |
High School Diploma/GED | 9.5% | 11.5% | |
Some College | 26.0% | 32.8% | |
Bachelor’s degree | 25.4% | 23.8% | |
Post-Graduate Degree | 34.1% | 28.7% | |
Household Income | M = $97,340, SD = $62,231 | M = $65,568 SD = $30,769 | |
Marital Status | 73.7% married/living together | 69.9% married/living together |
Note. M = Mean; SD = Standard deviation.
fMRI Paradigms and Processing
fMRI paradigms and processing are briefly summarized below. Details are provided in the supplemental materials.
Emotional Reactivity Paradigm: Emotional N-Back
In both studies, amygdalar response to positive and negative emotional stimuli was assessed during the Emotional N-back (EN-back) task (40, 41). EN-back is a block-designed fMRI paradigm with two imaging runs, including an active n-Back working memory component and a passive emotional perception component. Youths were instructed to complete the 0-back (i.e., responding “match” when the current stimulus matched a predetermined target) and 2-back (i.e., responding “match” when the current stimulus matched the one shown two trials before) tasks. The effects of the emotional stimuli were quantified during four blocks each of negative-facial (fearful), positive-facial (happy), neutral-facial, and non-facial (places) stimuli (42, 43). Behavioral performance accuracy was assessed through the percentage of correct trials.
fMRI Procedure
The ABCD study used three 3T scanner platforms (i.e., Siemens Prisma, General Electric 750, and Philips) across 21 sites with a harmonized neuroimaging protocol. Study I used tabulated and ROI-based results that were publicly shared (44, 45). Subcortical brain regions were defined with the FreeSurfer v5.3.0 atlas-based segmentation procedure (46). The quality control process followed the study-recommended criteria (45; see supplemental material for a detailed description). Accounting for 19.54% of participants whose fMRI data were not collected and 15.68% whose fMRI data were excluded, the final analyses included EN-back fMRI data from 67.14 % (n = 7,973) of youths.
In Study II, whole-brain fMRI was conducted using a GE 16-channel fixed-site Signa HDx 3.0 Tesla MRI scanner at the University of Georgia. FMRI data were acquired via a single-shot, gradient-echo echoplanar pulse sequence (TR = 2,000 ms; TE = 25 ms; FOV = 225 × 225 mm; matrix = 64 × 64 mm). Contiguous 3.5mm thick axial slices were acquired to provide whole-brain coverage in 3.5mm3 voxels. The left and right amygdala were defined by Talairach-space-based coordinates (x = ±23, y = 5, z = −15; 47) with a 5mm radius around each coordinator. Group summary maps were created to evaluate the validity of brain response to the emotional reactivity component of the EN-back task. In this study, EN-back fMRI data were obtained from 65.04% (n = 80) of participants, and data from 9.76% (n = 12) of participants were excluded in the quality control procedure (see supplemental material for a detailed description). The final analyses included EN-back fMRI data from 55.3% (n = 68) of adolescents.
In both studies, the neuroimaging data processing was conducted using the Analysis of Functional NeuroImages (AFNI) software (48). GLM was conducted with linear contrasts for positive/negative vs. neutral faces, reflecting youths’ emotional reactivity towards positive and negative facial expressions. GLM was conducted for each voxel of every participant using AFNI’s 3dDeconvolve (48), with the temporal pattern of the positive or negative emotion blocks and covariates (e.g., head movement values) as independent variables, neutral-face blocks as the baseline, and the blood-oxygen-level-dependent signal over time as the dependent variable. The average of resulting GLM beta coefficients of voxels in the left and right amygdala were computed across two runs for each participant, and were weighted by nominal degrees of freedom (44). See the supplemental material for more details.
Measures
Rearing Environments
In study I, rearing environments were assessed via family conflict (negative) and parental warmth (positive). Family conflict was measured via youth-reported Family Conflict subscale of the Family Environments Scale (FES; 49), and a sum of nine items was used in analyses (ω = .68). Parental warmth was measured via the shortened (5 items) Acceptance subscale from the Child Report of Parenting Behavior Inventory-Short version (CRPBI; 50). A sum score was used in the analyses (ω = .71).
In study II, rearing environments were indicated by family functioning assessed via caregiver-reported Family Adaptability and Cohesion Evaluation Scale (FACES) IV Package (51). FACES-IV measures family cohesion and flexibility dimensions comprehensively using 42 items from six scales: two balanced scales (i.e., cohesion and flexibility) and four unbalanced scales (i.e., rigid, chaotic, enmeshed, and disengaged). This study used the circumplex total ratio, which was calculated by dividing the average of the balanced scales by the average of the unbalanced scales with higher scores indicating more balanced family systems. The subscales presented acceptable internal consistency: cohesion (ω = .69), flexibility (ω = .67), rigid (ω = .63), chaotic (ω = .72), enmeshed (ω = .61), and disengaged (ω = .64).
Socio-emotional Outcomes
In both studies, youths’ internalizing & externalizing symptoms were assessed with caregiver-reported Child Behavior Checklist (CBCL; 52). Study I administered the CBCL at baseline and 1-year follow-up assessments. Raw scores of internalizing (ωStudy1 = .73, ωStudy2 = .80) and externalizing symptoms (ωStudy1 = .82, ωStudy2 = .87) were used in the analyses (52). For positive outcomes, Study I assessed youths’ prosocial behaviors at baseline and 1-year follow-up via the shortened (3 items; ω = .58) and caregiver-reported Prosocial Behavior subscale from the Strengths and Difficulties Questionnaire (53, 54). In Study II, youths’ empathy was assessed using the self-reported Perspective Taking and Emphatic Concern Scale (55), where the sum of 14 items (ω = .80) was analyzed.
Covariates
Both studies controlled for youths’ biological sex (“1” for female and “0” for male), age, race/ethnicity, household income, caregivers’ highest education, and marital status (“1” for married or living together and “0” for not married or living together). Education attainment was coded into an ordinal variable from “1” for less than high school diploma to “5” for postgraduate degrees. Household annual income was coded into an ordinal variable from “1” for less than $5,000 to “10” for more than $200,000. For Study I, youths’ race/ethnicity was coded into three binary variables: White, Black, and Hispanic/Latino(a). To retain statistical power in Study II, a minority status variable (“1” for minority and “0” for White/Caucasian) was controlled.
Analytical Plan
Hypotheses were tested using structural equation modeling (SEM) with maximum likelihood estimation with robust standard errors (56) in Mplus Version 8.3 (57). To examine the moderating effects of amygdalar reactivity on the associations between family environments and youths’ outcomes, interaction terms (i.e., centered amygdala reactivity × centered family environment) were added to the SEM models. The Benjamini-Hochberg procedure (58) was used to correct for multiple comparisons for each outcome variable across the four models in each study (false discovery rate = .05). Significant interaction effects that survived the multiple comparisons were further probed using the simple slope method (59, 60).
In Study I, we used two-level complex survey data (57) to correct for statistical dependencies related to nested participants in families and study sites. Sample propensity scores were used as weights to compensate for biases in the recruitment process. Little’s test (61) suggests missing-at-random for Study I (χ2(771) = 2712.73, p < .001) and missing-completely-at-random for Study II (χ2(70) = 66.23, p = .61). Thus, the full-information maximum likelihood (FIML) algorithm was employed to address missing data and include all participants (62). As a sensitivity analysis, we also analyzed all the models in both studies using listwise deletion, and the results were parallel with the main findings. Model fit was assessed using the chi-square test, the comparative fit index (CFI), and the standardized root mean square residual (SRMR).
Three critical tests were conducted as a rigorous examination of our hypothesis that amygdalar reactivity indicates youths’ differential sensitivity, for better and for worse, to rearing environments (32, 63). These tests included the region of significance (RoS), proportion of interaction (PoI), and proportion affected (PA) tests (see Fig1 for a visual presentation and detailed description). They aimed to differentiate BSCT/DST (i.e., differential sensitivity exhibits in both negative and positive environments) from diathesis-stress (i.e., differential sensitivity only shows in negative environments) and vantage sensitivity (i.e., differential sensitivity only shows in positive environments) hypotheses. BSCT/DST was supported only if 1) RoS showed on both positive & negative ends of environments, 2) PoI ranged 0.2–0.8, and 3) PA index ranged 16%–84%.
Fig1.
Visual Presentation of the Three Critical Tests to Differentiate BSCT/DST, Diathesis-Stress, and Vantage Sensitivity Models
Note. BSCT/DST is supported only if the model passes all three critical tests.
Results
Study variables were correlated in the expected directions (see Supplemental Tables 2&3 for correlation coefficients). Behavioral performance on the EN-back indicated that the included participants understood and performed well. In the full sample (before exclusion), the average accuracy rates were 74.78% for 2-back and 81.85% for 0-back in Study I, and 84.57% for 2-back and 90.82% for 0-back in Study II. After excluding low-performance youths, the average accuracy was 77.40% for 2-back and 85.31% for 0-back in Study I, and 88.33% for 2-back and 93.59% for 0-back in Study II. Group-level neural activation analyses associated with negative and positive emotional stimuli (presented in Supplemental Fig1) suggested significant group-level activation in the left and right amygdala in response to negative, but not positive, emotional stimuli (FWE p < .05).
In Study I (SEM model presented in Fig2; see Supplemental Tables 4&5 for model coefficients), left amygdalar response to positive stimuli significantly amplified the effect of parental warmth on youths’ prosocial behaviors (β = .03, p < .01) but not on psychopathology outcomes. According to interpretations (Fig3), the POI, PA, and RoS tests all supported the BSCT/DST hypothesis. Specifically, among participants with high left amygdalar response to positive stimuli, high parental warmth was associated with significantly increased prosocial behaviors (hypothesis 1), whereas low parental warmth was related to reduced prosocial behaviors (hypothesis 2). The effect of parental warmth on youths’ prosocial behaviors among those with low amygdalar reactivity was not significant (hypothesis 3). However, the moderating effects of the left amygdalar activation to negative stimuli and right amygdalar reactivity to positive/negative stimuli on the associations between rearing environments and youths’ outcomes were not significant.
Fig2.
Study I SEM Models Testing the Associations among Family Rearing Environments, Amygdalar Reactivity to Emotional Stimuli, and Youths’ Socio-Emotional Outcomes.
Note. T1 = Baseline assessments; T2 = 1-year follow-up; Solid lines indicate statistically significant associations while dotted lines represent insignificant associations. Standardized coefficients are presented in this figure. Covariate coefficients are omitted for figure clarity. *p < .05, **p < .01, *** p < .001.
Fig3.
Study I: The Interpretation of the Significant Moderation Effect of Left Amygdalar Reactivity to Positive Emotional Stimuli on the Associations Between Parental Warmth and Youths’ Prosocial Behaviors.
Note. The dashed line indicated that youths with low left amygdalar responses during positive emotional stimuli were not sensitive to the influences of parental warmth (n.s. = non-significant), while the solid lines suggested that for youths with high left amygdalar responses during positive emotional processing, parental warmth had significant impact on youths’ prosocial behaviors (***p < .001). PoI = Proportion of Interaction, with values between 0.2 and 0.8 suggesting potential BSCT/DST patterns. PA = Porportion affected, with values between 16% and 84% suggesting potential BSCT/DST patterns. Shaded regions (i.e, region of significance, RoS) indicated the values of parental warmth in which the moderator (i.e., amygdalar reactivity) significantly predicted the outcome (i.e., prosocial behaviors). BSCT/DST hypothese were supported because shaded regions appeared on both high and low extremes of the independent variable (i.e., parental warmth). This figure passed all the three tests (i.e., POI, PA, and RoS), and thus supported the BSCT/DST “for better and for worse” hypothesis.
In Study II (SEM model presented in Fig4; see Supplemental Tables 6 for model coefficients), left amygdalar response to positive and negative stimuli significantly amplified the influence of family functioning on youths’ internalizing (negative stimuli: β = −.33, p < .05; positive stimuli: β = −.38, p < .01) and externalizing (negative stimuli: β = −.32, p < .05; positive stimuli: β = −.32, p < .01) symptoms. Right amygdalar reactivity to positive stimuli significantly exacerbated the associations between family functioning and internalizing (β = −.30, p < .01) but not externalizing problems (β = −.23, p = .06). Right amygdalar response to negative stimuli did not significantly moderate the links between family functioning and youths’ outcomes. No significant interaction effect on youths’ empathy outcomes was observed.
Fig4.
Study II SEM Models Testing the Associations among Family Functioning, Amygdalar Activation to Emotional Stimuli, and Youths’ Socio-Emotional Outcomes.
Note. Solid lines indicate statistically significant associations while dotted lines represent insignificant associations. Standardized coefficients are presented in this figure. Covariate coefficients are omitted for figure clarity. *p < .05, **p < .01, *** p < .001.
Fig5 shows Study II results that supported BSCT/DST “for better and for worse” hypothesis as indicated by the POI, PA, and RoS tests. Compared to youths with low amygdalar reactivity, youths with high left (Panel-A) or right (Panel-B) amygdalar response to positive stimuli exhibited elevated internalizing problems under the influence of unbalanced family functioning. These youths also presented fewer internalizing symptoms when exposed to balanced family functioning. Fig6 shows results that are consistent with the diathesis-stress model (22). Compared to youths with low amygdalar reactivity, youths with high left amygdalar reactivity to negative stimuli showed elevated externalizing (Panel-A) and internalizing (Panel-C) symptoms under the impact of unbalanced family functioning. Similarly, youths with high left amygdalar response to positive stimuli exhibited elevated externalizing problems (Panel-B) in environments with unbalanced family functioning. However, these effects presented in Fig6 failed the RoS on X test, suggesting that youths did not significantly benefit from balanced family functioning. Effects probed in Panel (B) & (C) also failed the PoI test. Therefore, these effects were aligned with the diathesis-stress hypothesis.
Fig5.
Study II: The Interpretation of the Significant Moderation Effects Consistent with the BSCT/DST Hypotheses.
Note. Panel (A) presents the plot of the moderating effect of left amygdalar reactivity to positive emotional stimuli on the links between family functioning and internalizing symptoms; Panel (B) presents the plot of the moderating effect of right amygdalar reactivity to positive emotional stimuli on the links between family functioning and internalizing symptoms. The dashed line indicated that youths with low left/right amygdalar responses during positive emotional processing were not sensitive to the influences of family functioning (n.s. = non-significant), while the solid lines suggested that for youths with high left/right amygdalar responses during positive emotional stimuli, family functioning had significant impact on their internalzing problems (**p < .01, ***p < .001). PoI = Proportion of interaction, with values between 0.2 and 0.8 suggesting potential BSCT/DST patterns. PA = Porportion affected, with values between 16% and 84% suggesting potential BSCT/DST patterns. Shaded regions (i.e, region of significance, RoS) indicated the values of family functioning in which the moderator (i.e., amygdalar reactivity) significantly predicted the outcome (i.e., internalizing & externalizing problems). BSCT/DST hypothese were supported when shaded regions appeared on both high and low extremes of the independent variable (i.e., family functioning). These two figures passed all the three tests (i.e., POI, PA, and RoS), and thus supported the BSCT/DST “for better and for worse” hypothesis.
Fig6.
Study II: The Interpretation of the Significant Moderation Effects Consistent with the Diathesis-Stress Model.
Note. Panel (A) presents the plot of the moderating effect of left amygdalar reactivity to negative emotional stimuli on the links between family functioning and externalizing symptoms; Panel (B) presents the plot of the moderating effect of left amygdalar reactivity to positive emotional stimuli on the links between family functioning and externalizing symptoms; Panel (C) presents the plot of the moderating effect of left amygdalar reactivity to negative emotional stimuli on the links between family functioning and internalizing symptoms. The dashed line indicated that youths with low left amygdalar responses during negative/positive emotional stimuli were not sensitive to the influences of family functioning (n.s. = non-significant), while the solid lines suggested that for youths with high left amygdalar responses during negative/positive emotional stimuli, family functioning had significant impact on internalzing/externalizing problems (*p < .05, **p < .01, ***p < .001). PoI = Proportion of interaction, with values between 0.2 and 0.8 suggesting potential BSCT/DST patterns. PA = Porportion affected, with values between 16% and 84% suggesting potential BSCT/DST patterns. Shaded regions (i.e, region of significance, RoS) indicated the values of family functioning in which the moderator (i.e., amygdalar reactivity) significantly predicted the outcome (i.e., internalizing & externalizing problems). BSCT/DST hypothese were supported when shaded regions appeared on both high and low extremes of the independent variable (i.e., family functioning). In these three figures, because amygdalar reactivity only significantly predicted chagnes in youths’ outcomes on the unbalanced extreme of family functioning (i.e., RoS only appeared on the extreme of unbalanced family functioning), these results were consistent with the diathesis-stress model.
Discussion
We used two studies to test whether amygdalar reactivity to emotional stimuli was a differential sensitivity biomarker that moderated the effects of family environments on youths’ socio-emotional adjustment. Findings partially supported BSCT/DST (5–7). Results from Study I showed that left amygdalar response to positive stimuli significantly intensified the effects of low/high parental warmth on youths’ decreased/increased prosocial behaviors. In Study II, left/right amygdala reactivity to positive stimuli significantly amplified the associations between unbalanced/balanced family functioning and youths’ elevated/reduced internalizing problems. Additionally, findings from Study II partially evidenced support for the diathesis-stress hypothesis (22). Left amygdalar response to negative stimuli amplified the influences of unbalanced family functioning on youths’ heightened internalizing and externalizing symptoms. Similar risk-potentiation effects were also found for left amygdalar reactivity to positive stimuli but only for elevated externalizing problems.
The findings of amygdalar reactivity as a differential sensitivity indicator aligns with amygdalar functions in the perception and processing of emotional stimuli. Specifically, the amygdala partakes an important role in associating innate and learned emotional salience to percepts and influences downstream emotional reactivity (64). These associations emerge via adverse or supportive caregiving experiences and subsequently facilitate youths to initiate corresponding behavioral reactions (65, 66). Therefore, greater amygdalar response during emotional processing is associated with elevated sensitivity to positive and/or negative environments, while blunted amygdalar reactivity is linked to youths’ reduced responsivity to family influences. These findings corroborated and extended previous empirical evidence from young-adult samples (30, 31) and demonstrated the role of amygdalar reactivity as a differential sensitivity indicator across developmental stages.
Some differences existed between the findings of the two studies. In Study I, amygdalar reactivity only potentiated supportive environmental effects on positive youth outcomes. But in Study II, amygdalar reactivity potentiated both supportive and adverse family environmental influences on negative youth outcomes. This distinction might be related to the methodological differences between these two studies. In particular, Study I used longitudinal data and assessed the change in youths’ behavioral outcomes. Given that prosocial behaviors (β < .50) showed less stability than psychopathology outcomes (β > .60), it was possible that prosocial behaviors were more malleable to change. In contrast, Study II used cross-sectional data that captured youths’ psychopathology and empathy levels. The significant results in relation to psychopathology outcomes, not empathy, could be explained by the larger variances observed in the psychopathology variables. The second difference was that partial findings from Study II, but not Study I, supported the diathesis-stress hypothesis. Because youths in Study II were of lower SES and showed higher mean levels of psychopathology, they might be more sensitive to negative environmental influences and thus present vulnerability patterns consistent with the diathesis-stress model. For these at-risk youths in Study II, the positive effects of balanced family functioning may be weaker, and thereby failing to differentiate high-sensitivity youths’ psychopathology symptoms from those with low sensitivity.
In this investigation, the majority of significant differential sensitivity effects were found to be associated with the left amygdalar reactivity. The laterality differences of amygdalar temporal dynamics may contribute to the laterality differences of amygdalar functions in differential sensitivity (34). Accordingly, the right amygdala is involved in the rapid and automatic detection of emotional stimuli, while the left amygdala mediates the more detailed and elaborate analyses of positive and negative emotions (67, 68). In this study, youths’ emotional reactivity was elicited during challenging n-back cognitive tasks that involved a range of emotional facial expressions. Thus, the left amygdalar reactivity during EN-back reflected youths’ ability to evaluate detailed and specific emotional information while they were cognitively engaged. This ability may play a more salient role in determining the emotional characteristics of the rearing environments and generate behavioral reactions accordingly (67). It is possible that the left amygdala is more critical for shaping youths’ sensitivity to rearing environments.
Limitations
The current investigation has several limitations. First, amygdalar reactivity has limited longitudinal stability during adolescence. A recent meta-analysis found low intraclass correlation coefficients and suggested limited test-retest reliability of task-based fMRI measures in individual-differences research (69). Thus, amygdalar reactivity may serve as a transient sensitivity indicator that is subject to change during adolescence (70). Future longitudinal studies are needed to reveal the trajectory of changes in differential sensitivity biomarkers across developmental stages. For Study I, the shortened or trimmed measurement tools might lead to compromised construct validity. Caregiverreported prosocial behaviors also presented relatively low reliability (ω = .58). Additionally, family conflict and parental warmth were not mutually-exclusive and did not reflect the full spectrum of rearing environments (71), which is not ideal for testing BSCT/DST. Lastly, the propensity weights used in analyses only adjusted for sampling biases but did not take fMRI data missingness into consideration. In Study II, families were recruited from rural Georgia, thus limiting the generalizability of the findings to urban areas and other regions of the U.S. Because of the cross-sectional data in Study II, the temporal precedence among study variables could not be determined.
Implications
The findings of this investigation extended BSCT/DST literature by revealing amygdalar reactivity to emotional stimuli as a neural process underlying youths’ differential sensitivity to rearing environments (18). In practice, research on the structure, function, and connectivity of specific brain regions (e.g., amygdala) and other sensitivity indicators will help establish a set of neurobiological sensitivity biomarkers. This set of sensitivity biomarkers can help intervention researchers identify at-risk youths with more or less sensitivity to rearing environments and tailor intervention practices accordingly (23). Preventive intervention programs may also use neurobiological sensitivity indicators to monitor progress during and after intervention programs (23). Further, because of the significant brain changes during adolescence (72), amygdalar reactivity is a transient sensitivity indicator subject to change over time and under environmental influences. This neural plasticity enables youths with the potential to reprogram their neural functions to be consistent with current experiences (23, 72).
Supplementary Material
Acknowledgments
Data used in the preparation of Study I of this article were obtained from the Adolescent Brain Cognitive DevelopmentSM (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 age 9-10 and follow them over ten years into early adulthood. The ABCD Study® is supported by the National Institutes of Health and additional federal partners under award numbers U01DA041048, U01DA050989, U01DA051016, U01DA041022, U01DA051018, U01DA051037, U01DA050987, U01DA041174, U01DA041106, U01DA041117, U01DA041028, U01DA041134, U01DA050988, U01DA051039, U01DA041156, U01DA041025, U01DA041120, U01DA051038, U01DA041148, U01DA041093, U01DA041089, U24DA041123, U24DA041147. A full list of supporters is available at https://abcdstudy.org/federal-partners.html. A listing of participating sites and a complete listing of the study investigators can be found at https://abcdstudy.org/consortium_members/. ABCD consortium investigators designed and implemented the study and/or provided data but did not necessarily participate in the analysis or writing of this report. The ABCD data repository grows and changes over time. The ABCD data used in this report came from curated annual release 2.0.1. Data used in the preparation of Study II of this article were supported by Award K01DA045219 (principle investigator: Assaf Oshri) by the National Institute on Drug Abuse (NIDA). This manuscript reflects the views of the author and may not reflect the opinions or views of the NIH or ABCD consortium investigators.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures
The authors declare no biomedical financial interests or potential conflicts of interest.
References
- 1.Rohner RP, Khaleque A (2010): Testing central postulates of parental acceptance-rejection theory (PARTheory): A meta-analysis of cross-cultural studies. Journal of Family Theory & Review. 2:73–87. [Google Scholar]
- 2.Kim-Spoon J, Cicchetti D, Rogosch FA (2013): A longitudinal study of emotion regulation, emotion lability-negativity, and internalizing symptomatology in maltreated and nonmaltreated children. Child development. 84:512–527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Masten AS, Obradović J (2006): Competence and resilience in development. Annals of the New York Academy of Sciences. 1094:13–27. [DOI] [PubMed] [Google Scholar]
- 4.Luthar SS (2006): Resilience in development: A synthesis of research across five decades.
- 5.Boyce WT, Ellis BJ (2005): Biological sensitivity to context: I. An evolutionary–developmental theory of the origins and functions of stress reactivity. Development and psychopathology. 17:271–301. [DOI] [PubMed] [Google Scholar]
- 6.Ellis BJ, Essex MJ, Boyce WT (2005): Biological sensitivity to context: II. Empirical explorations of an evolutionary–developmental theory. Development and psychopathology. 17:303–328. [DOI] [PubMed] [Google Scholar]
- 7.Belsky J, Pluess M (2009): Beyond diathesis stress: differential susceptibility to environmental influences. Psychological bulletin. 135:885. [DOI] [PubMed] [Google Scholar]
- 8.Guyer AE (2020): Adolescent Psychopathology: The Role of Brain-Based Diatheses, Sensitivities, and Susceptibilities. Child Development Perspectives. 14:104–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Morris AS, Silk JS, Steinberg L, Myers SS, Robinson LR (2007): The role of the family context in the development of emotion regulation. Social development. 16:361–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sheeber L, Hops H, Alpert A, Davis B, Andrews J (1997): Family support and conflict: Prospective relations to adolescent depression. Journal of abnormal child psychology. 25:333–344. [DOI] [PubMed] [Google Scholar]
- 11.Davies PT, Harold GT, Goeke-Morey MC, Cummings EM, Shelton K, Rasi JA, et al. (2002): Child emotional security and interparental conflict. Monographs of the society for research in child development.i-127. [PubMed] [Google Scholar]
- 12.Padilla-Walker LM, Nielson MG, Day RD (2016): The role of parental warmth and hostility on adolescents’ prosocial behavior toward multiple targets. Journal of Family Psychology. 30:331. [DOI] [PubMed] [Google Scholar]
- 13.Olson DH (2000): Circumplex model of marital and family systems. Journal of family therapy. 22:144–167. [Google Scholar]
- 14.Lewandowski AS, Palermo TM, Stinson J, Handley S, Chambers CT (2010): Systematic review of family functioning in families of children and adolescents with chronic pain. The Journal of Pain. 11:1027–1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Olson DH (2000): Circumplex model of marital and family sytems. Journal of family therapy. 22:144–167. [Google Scholar]
- 16.Mastrotheodoros S, Canário C, Gugliandolo MC, Merkas M, Keijsers L (2019): Family functioning and adolescent internalizing and externalizing problems: Disentangling between-, and within-family associations. Journal of youth and adolescence.1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Renzaho A, Mellor D, McCabe M, Powell M (2013): Family functioning, parental psychological distress and child behaviours: Evidence from the Victorian child health and wellbeing study. Australian Psychologist. 48:217–225. [Google Scholar]
- 18.Ellis BJ, Boyce WT, Belsky J, Bakermans-Kranenburg MJ, Van IJzendoorn MH (2011): Differential susceptibility to the environment: An evolutionary–neurodevelopmental theory. Development and psychopathology. 23:7–28. [DOI] [PubMed] [Google Scholar]
- 19.Slagt M, Dubas JS, Ellis BJ, Van Aken MA, Deković M (2019): Linking emotional reactivity “for better and for worse” to differential susceptibility to parenting among kindergartners. Development and psychopathology. 31:741–758. [DOI] [PubMed] [Google Scholar]
- 20.Slagt M, Dubas JS, Deković M, van Aken MA (2016): Differences in sensitivity to parenting depending on child temperament: A meta-analysis. Psychological bulletin. 142:1068. [DOI] [PubMed] [Google Scholar]
- 21.Pluess M, Belsky J (2013): Vantage sensitivity: Individual differences in response to positive experiences. Psychological bulletin. 139:901. [DOI] [PubMed] [Google Scholar]
- 22.Monroe SM, Simons AD (1991): Diathesis-stress theories in the context of life stress research: implications for the depressive disorders. Psychological bulletin. 110:406. [DOI] [PubMed] [Google Scholar]
- 23.Schriber RA, Guyer AE (2016): Adolescent neurobiological susceptibility to social context. Developmental cognitive neuroscience. 19:1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Telzer EH, Jorgensen NA, Prinstein MJ, Lindquist KA (2020): Neurobiological sensitivity to social rewards and punishments moderates link between peer norms and adolescent risk taking. Child development. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rudolph KD, Davis MM, Modi HH, Fowler C, Kim Y, Telzer EH (2020): Differential susceptibility to parenting in adolescent girls: moderation by neural sensitivity to social cues. Journal of Research on Adolescence. 30:177–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.LeDoux JE (2000): Emotion circuits in the brain. Annual review of neuroscience. 23:155–184. [DOI] [PubMed] [Google Scholar]
- 27.Garavan H, Pendergrass JC, Ross TJ, Stein EA, Risinger RC (2001): Amygdala response to both positively and negatively valenced stimuli. Neuroreport. 12:2779–2783. [DOI] [PubMed] [Google Scholar]
- 28.Davis M, Whalen PJ (2001): The amygdala: vigilance and emotion. Molecular psychiatry. 6:13. [DOI] [PubMed] [Google Scholar]
- 29.Herman JP, Ostrander MM, Mueller NK, Figueiredo H (2005): Limbic system mechanisms of stress regulation: hypothalamo-pituitary-adrenocortical axis. Progress in Neuro-Psychopharmacology and Biological Psychiatry. 29:1201–1213. [DOI] [PubMed] [Google Scholar]
- 30.Gard AM, Shaw DS, Forbes EE, Hyde LW (2018): Amygdala reactivity as a marker of differential susceptibility to socioeconomic resources during early adulthood. Developmental psychology. 54:2341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Swartz JR, Knodt AR, Radtke SR, Hariri AR (2015): A neural biomarker of psychological vulnerability to future life stress. Neuron. 85:505–511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Roisman GI, Newman DA, Fraley RC, Haltigan JD, Groh AM, Haydon KC (2012): Distinguishing differential susceptibility from diathesis–stress: Recommendations for evaluating interaction effects. Development and psychopathology. 24:389–409. [DOI] [PubMed] [Google Scholar]
- 33.Widaman KF, Helm JL, Castro-Schilo L, Pluess M, Stallings MC, Belsky J (2012): Distinguishing ordinal and disordinal interactions. Psychological methods. 17:615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sergerie K, Chochol C, Armony JL (2008): The role of the amygdala in emotional processing: a quantitative meta-analysis of functional neuroimaging studies. Neuroscience & Biobehavioral Reviews. 32:811–830. [DOI] [PubMed] [Google Scholar]
- 35.Baas D, Aleman A, Kahn RS (2004): Lateralization of amygdala activation: a systematic review of functional neuroimaging studies. Brain Research Reviews. 45:96–103. [DOI] [PubMed] [Google Scholar]
- 36.Fusar-Poli P, Placentino A, Carletti F, Landi P, Allen P, Surguladze S, et al. (2009): Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. Journal of psychiatry & neuroscience. [PMC free article] [PubMed] [Google Scholar]
- 37.Garavan H, Bartsch H, Conway K, Decastro A, Goldstein R, Heeringa S, et al. (2018): Recruiting the ABCD sample: design considerations and procedures. Developmental cognitive neuroscience. 32:16–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Volkow ND, Koob GF, Croyle RT, Bianchi DW, Gordon JA, Koroshetz WJ, et al. (2018): The conception of the ABCD study: From substance use to a broad NIH collaboration. Developmental cognitive neuroscience. 32:4–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Rosenbaum PR, Rubin DB (1983): The central role of the propensity score in observational studies for causal effects. Biometrika. 70:41–55. [Google Scholar]
- 40.Cohen A, Conley M, Dellarco D, Casey B (2016): The impact of emotional cues on short-term and long-term memory during adolescence. Proceedings of the Society for Neuroscience San Diego, CA November. [Google Scholar]
- 41.Barch DM, Burgess GC, Harms MP, Petersen SE, Schlaggar BL, Corbetta M, et al. (2013): Function in the human connectome: task-fMRI and individual differences in behavior. Neuroimage. 80:169–189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Conley MI, Dellarco DV, Rubien-Thomas E, Cohen AO, Cervera A, Tottenham N, et al. (2018): The racially diverse affective expression (RADIATE) face stimulus set. Psychiatry research. 270:1059–1067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Tottenham N, Tanaka JW, Leon AC, McCarry T, Nurse M, Hare TA, et al. (2009): The NimStim set of facial expressions: judgments from untrained research participants. Psychiatry research. 168:242–249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Hagler DJ Jr, Hatton S, Cornejo MD, Makowski C, Fair DA, Dick AS, et al. (2019): Image processing and analysis methods for the Adolescent Brain Cognitive Development Study. NeuroImage.116091. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Casey BJ, Cannonier T, Conley MI, Cohen AO, Barch DM, Heitzeg MM, et al. (2018): The adolescent brain cognitive development (ABCD) study: imaging acquisition across 21 sites. Developmental cognitive neuroscience. 32:43–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, et al. (2002): Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 33:341–355. [DOI] [PubMed] [Google Scholar]
- 47.Talairach J, Tournoux P (1988): Co-planar stereotaxic atlas of the human brain. 1988. New York: Theime. [Google Scholar]
- 48.Cox RW (1996): AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical research. 29:162–173. [DOI] [PubMed] [Google Scholar]
- 49.Moos RH, Moos BS (1994): Family environment scale manual. Consulting Psychologists Press. [Google Scholar]
- 50.Schaefer ES (1965): A configurational analysis of children’s reports of parent behavior. Journal of consulting psychology. 29:552. [DOI] [PubMed] [Google Scholar]
- 51.Olson DH, Gorall D, Tiesel J (2006): FACES IV Package. Minneapolis, MN: Life Innovations. Inc. [Google Scholar]
- 52.Achenbach TM, Dumenci L, Rescorla LA (2001): Ratings of relations between DSM-IV diagnostic categories and items of the CBCL/6–18, TRF, and YSR. Burlington, VT: University of Vermont. [Google Scholar]
- 53.Goodman R, Meltzer H, Bailey V (1998): The Strengths and Difficulties Questionnaire: A pilot study on the validity of the self-report version. European child & adolescent psychiatry. 7:125–130. [DOI] [PubMed] [Google Scholar]
- 54.Goodman R, Scott S (1999): Comparing the Strengths and Difficulties Questionnaire and the Child Behavior Checklist: is small beautiful? Journal of abnormal child psychology. 27:17–24. [DOI] [PubMed] [Google Scholar]
- 55.Davis MH (1983): Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of personality & social psychology. 44:113. [Google Scholar]
- 56.Yuan K-H, Bentler PM (2000): 5. Three likelihood-based methods for mean and covariance structure analysis with nonnormal missing data. Sociological methodology. 30:165–200. [Google Scholar]
- 57.Muthén L, Muthén B (2012): Mplus user’s guide (1998–2012). Los Angeles, CA: Muthén & Muthén. 6. [Google Scholar]
- 58.Benjamini Y, Hochberg Y (1995): Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological). 57:289–300. [Google Scholar]
- 59.Aiken LS, West SG, Reno RR (1991): Multiple regression: Testing and interpreting interactions. Sage. [Google Scholar]
- 60.Dawson JF (2014): Moderation in management research: What, why, when, and how. Journal of Business and Psychology. 29:1–19. [Google Scholar]
- 61.Little RJ, Rubin D (2002): Statistical analysis with missing data. Wiley. New York. [Google Scholar]
- 62.Little RJ, Rubin DB (2019): Statistical analysis with missing data. John Wiley & Sons. [Google Scholar]
- 63.Del Giudice M (2017): Statistical tests of differential susceptibility: Performance, limitations, and improvements. Development and psychopathology. 29:1267–1278. [DOI] [PubMed] [Google Scholar]
- 64.Hamann SB, Mao H (2002): Positive and negative emotional verbal stimuli elicit activity in the left amygdala. Neuroreport. 13:15–19. [DOI] [PubMed] [Google Scholar]
- 65.Windle M, Gray JC, Lei KM, Barton AW, Brody G, Beach SR, et al. (2018): Age sensitive associations of adolescent substance use with amygdalar, ventral striatum, and frontal volumes in young adulthood. Drug and alcohol dependence. 186:94–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Hanson JL, Nacewicz BM, Sutterer MJ, Cayo AA, Schaefer SM, Rudolph KD, et al. (2015): Behavioral problems after early life stress: contributions of the hippocampus and amygdala. Biological psychiatry. 77:314–323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Wright CI, Martis B, Schwartz CE, Shin LM, åkan Fischer H, McMullin K, et al. (2003): Novelty responses and differential effects of order in the amygdala, substantia innominata, and inferior temporal cortex. Neuroimage. 18:660–669. [DOI] [PubMed] [Google Scholar]
- 68.Gläscher J, Adolphs R (2003): Processing of the arousal of subliminal and supraliminal emotional stimuli by the human amygdala. Journal of Neuroscience. 23:10274–10282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69.Elliott ML, Knodt AR, Ireland D, Morris ML, Poulton R, Ramrakha S, et al. (2020): What is the test-retest reliability of common task-functional MRI measures? New empirical evidence and a meta-analysis. Psychological Science. 31:792–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Manuck S (2011): Species of gene–environment interaction: Diathesis–stress, vantage sensitivity, and differential susceptibility. Carolina Consortium on Human Development. [Google Scholar]
- 71.Skopp NA, McDonald R, Jouriles EN, Rosenfield D (2007): Partner aggression and children’s externalizing problems: Maternal and partner warmth as protective factors. Journal of Family Psychology. 21:459. [DOI] [PubMed] [Google Scholar]
- 72.Bredy TW, Zhang TY, Grant RJ, Diorio J, Meaney MJ (2004): Peripubertal environmental enrichment reverses the effects of maternal care on hippocampal development and glutamate receptor subunit expression. European Journal of Neuroscience. 20:1355–1362. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.