Sad, Sadder, Saddest: Recognition of Sad and Happy Emotional Intensity, Adverse Childhood Experiences and Depressive Symptoms in Preschoolers

Abstract

Adverse childhood experiences (ACES) have repeatedly been associated with depression. The ability to differentiate emotional intensity is a protective factor for psychopathology and in the context of life stressors, poor negative emotion differentiation (ED) is associated with depressive symptoms. However, little is known about whether the ability to recognize negative emotional intensity, a theorized developmental prerequisite of ED, influences the relationship between ACES and depressive symptoms in early childhood. The current study examined the interactive effects of ACES, the ability to recognize emotional intensity and depressive symptoms in 249 preschoolers enriched for depression. Findings demonstrated that when experiencing ACES, sad (not happy) emotion recognition was associated with elevated depressive symptoms. Specifically, when facing multiple ACEs, preschoolers with poor and moderate ability to recognize sad emotional intensity exhibited elevated depressive symptoms. Findings demonstrate that when experiencing elevated ACES, sad emotion recognition may be a protective factor for depression in early childhood.

The ability to recognize emotions, one form of emotional competence, is vital in our day-to-day functioning as it is used to inform judgement, processing, cognition and emotion state [1]. This ability is important in overall well-being as well as interpersonal success [2]. One key component of emotional competence is the ability to recognize emotions and then to differentiate them from other emotions, including those of the same valence. Emotion recognition, or the ability to recognize discrete emotions, is an early step in emotional competence and is a purported developmental precursor for the ability to differentiate emotions more broadly. Emotion Differentiation is defined as one’s ability to differentiate affective experiences into distinct categories (e.g., anger vs. fear) [3], including emotions that are distinct within the same valence [4]. A better ability to differentiate negative emotions has repeatedly been associated with emotional well-being [5–9] and better psychological adjustment [3]. Findings generally indicate stronger associations of negative emotion differentiation (e.g., differentiating sadness vs. frustration) than positive emotion differentiation (e.g., contentment vs. excitement) with psychopathology outcomes [10–12] and in particular, negative emotion differentiation has been implicated in depression [11]. Greater depressive symptoms are associated with lower negative emotion differentiation [11] and conversely the ability to differentiate negative emotions is negatively correlated with depressive symptoms in adulthood [13].

Much of the above literature pertains to adults, with less work examining aspects of emotional competence earlier in development, and in particular, during the preschool age, a period when these skills are developing. Emotion differentiation develops systemically, such that children start with broad categories of emotion based on valence and continue to differentiate them into more nuanced adult emotional categories [14]. Children as young as 3.5 years appear to be able to categorize emotions along the happy-sad continuum as well as adults [15] and have been found to identify and label emotions of happiness and sadness as accurately as undergraduates [16]. More complex processes necessary to aid in emotion differentiation and recognition, such as emotional comprehension and emotional appraisal, appear to develop and improve from age 4 to age 5 [17], which is parallel to the development of complex cognitive processes including theory of mind. Taken together, emotional understanding develops as early as preschool and from there can inform children’s understanding of other’s thinking or theory of mind [18].

Beyond cognition, children as young as preschool have been found to react in physiologically different ways to distinct negative emotions, and this ability to discriminate negative emotions increases from preschool to second grade [19]. Five year-old children’s ability to judge trustworthiness of faces in a socio-emotional context is also related to their development of emotional comprehension [20]. Emotional expression and regulation are also found to be associated with school adjustment [21]. Even in non-social environments, children’s reported understanding and experience of egocentric mixed emotions is evident as young as 6 years old and increases through development [22]. The ability to recognize more subtle emotional expressions suggests a keener recognition ability and therefore is considered a developmental precursor to emotion differentiation. Indeed, an ability to recognize emotions is associated with similar adaptive outcomes as emotion differentiation, including better well-being and the initiation and maintenance of healthy social relationships [23], as well as deficits implicated in psychopathology, in particular depression [2]. Importantly, emotion recognition deficits have also been found to be a trans-diagnostic risk factor across multiple disorders, especially in children [24]. In order to measure this nuanced form of emotion recognition early in life, research often utilizes developmentally appropriate adaptations [25–26] rather than the standard ecological momentary assessment techniques most commonly used in adults [27].

The current study uses a task that asks children to recognize differences in emotional intensity between two happy faces or between two sad faces [25–26]. This measure of emotion recognition uses prototypical negative and positive emotions that are developmentally appropriate (i.e., happy and sad). An adaptation of this task for preschool aged children was developed in collaboration with Gur and colleagues [28] and elicits a more fine-grained assessment of the child’s ability to assess and differentiate emotional intensity between two sad faces. The recognition of emotion intensity represents a method to assess more subtle emotion expressions that suggests enhanced recognition ability above simple emotion identification. Further, initial evidence indicates this form of emotional recognition is an early-emerging presentation of emotion differentiation that continues to develop into the full emotion differentiation ability through adulthood [29]. Using the same preschool sample as the current study, individual differences on this emotion recognition task predict elevated body max index (BMI) percentiles across development and peer relations in preschool and moderated the relationship between preschool emotional recognition ability and adolescent BMI [30–31]. Together, the ability to recognize differing emotional intensity is evident in early childhood, yet has not fully been explored in developmental populations in relation to individual psychiatric outcomes.

Of the little research examining emotion recognition in preschoolers, most studies have focused on basic emotion recognition and without consideration of the effects of psychopathology, including depression. Research on emotion recognition in childhood (i.e., simply identifying emotions rather than a more nuanced recognition of differing emotional intensity), has found a modest, negative correlation between positive emotion recognition ability and depressive symptoms in childhood and adolescent twins ages 9 to 17 [32]. What has been done so far on more nuanced emotion recognition has focused on older samples of adolescents and young adults, which therefore has studied a more fully developed form of emotion differentiation. Finding suggest that similar to adults, young adults with low negative emotion differentiation demonstrate stronger associations between daily brooding and depressive symptoms [33]. Furthermore, adolescents with low negative emotion differentiation were also found to exhibit elevated depressive symptoms, specifically in the context of high stress exposure [34]. Overall, lower negative emotion differentiation has been significantly linked to depression via moderating effects and increasing major depressive disorder symptom severity while high negative emotion differentiation appears to be adaptive in major depressive disorder. Though there is initial work in adolescents suggesting similar relations, little is known of whether the emerging ability to recognize and differentiate emotions is associated with risk for depression in early childhood, a key period of rapid emotional development.

Another important factor that has also been associated with risk for psychopathology in children is exposure to adverse childhood experiences or ACEs [35]. ACEs are defined as traumatic events that occur in childhood and include aspects of the child’s environment that can affect their sense of safety, stability, and attachment [36]. According to the Center for Disease Control and Prevention (CDC), these can include physical, emotional, or sexual abuse, physical or emotional neglect, or household dysfunction including mental illness, incarcerated relatives, mother treated violently, substance abuse, or divorce. ACEs has been found to be associated with maladaptive outcomes at multiple stages of life. From childhood to young adulthood, experience of ACEs predicts lower grade point averages in students as well as increased maladaptive behaviors including poor school attendance, behavioral problems, and failure to meet grade-level standards [37–38]. During the preschool years, exposure to multiple forms of ACEs significantly increase the likelihood of physical, mental, and developmental health conditions [39]. Moreover, preschoolers exposed to maltreatment exhibited significant deficits in executive function [40], developmental delays, and behavioral problems [41] while children as young as 6 years old who experience multiple forms of ACEs are more likely to have internalizing disorders compared to those experiencing no ACEs [42]. One facet of ACEs in particular, maltreatment, demonstrates strong associations with major depressive disorder and post-traumatic stress symptoms in children as young as 3–5 years [43]. In adulthood, individuals who experienced ACEs were more likely to binge drink compared to those without those experiences [44]. In relation to risk for psychopathology, ACEs predict worsening of mental health over the course of adolescence to adulthood, including depressive symptoms, antisocial behavior, and drug use [45]. ACEs has also been connected to other mental disorders including PTSD, such that early childhood onset of ACEs significantly predicted elevated PTSD and depression symptoms in pregnant women [46].

Exposure to ACEs is associated with increased risk of depressive disorders in adulthood, decades after their occurrence [47]. Indeed, emotion differentiation has been shown to be an important moderator of depressive outcomes in the context of exposure to stressful life experiences [48]. Low negative emotion differentiation is depressogenic in the context of high stress exposure in adolescence. In particular, Starr et. al. [48] provides evidence of the moderating role of emotion differentiation between stressful life experiences and depression in adolescence. Little is still known about early emotion recognition, as a moderator between severe stressful life events such as ACEs and depression, especially in preschool-aged populations, when emotional competence is rapidly developing.

The current study focused on examining the effects of ACEs, emotion recognition and depression in early childhood. Building on the extant literature in adult and adolescent samples, we investigate how mental health outcomes relate concurrently with ACEs and whether emotion recognition ability is associated with this relationship. The current study hypothesized that in the context of ACEs, low negative (sad) emotion recognition ability would be associated with depressive symptoms in preschool-aged children. Given the mixed literature on the relationship with positive emotion differentiation and depression in older populations, we did not hypothesize relationships between ACEs, positive (happy) emotion recognition, and depressive symptoms.

Methods

Participants were from the Preschool Depression Study (PDS), a prospective, longitudinal investigation at the Washington University School of Medicine (WUSM), examining 306 preschoolers and their families. Preschoolers between 3 and 6 years old were recruited from the St. Louis area through pediatrician offices, primary care practices, preschool and daycare centers, and recruitment sites accessible to the general public regardless of socio-economic status in order to create a stratified sample. The sample was enriched for preschoolers with early onset depressive symptoms using the Preschool Feelings Checklist (PFC), a validated measure for identifying children at high risk for preschool-onset depression [49–50]. In addition to preschoolers with high symptom scores, children with low or no endorsed symptoms were all also recruited and asked to participate. Children with chronic medical illnesses, neurological problems, pervasive developmental disorders, or language and/or cognitive delays as well as those outside the age range for the study were excluded. Exclusionary criteria was assessed via caregiver report during an initial phone screen, with caregivers providing any relevant diagnoses, known delays, and indication of impairment in any of the abovementioned areas. Parental written consent and child assent were obtained prior to participation and the WUSM Institutional Review Board approved all procedures. The current study examined 249 preschoolers in the PDS sample who completed the emotion recognition task and the measure of ACEs at baseline (mean age (SD)=5.43(.85).

Measures

Emotion Recognition

The current study used the 40-item version of the Penn Emotion Differentiation Test [25–26] as a measure of emerging emotion differentiation or emotion intensity recognition of happy and sad emotions. At Time 1, each participant was shown 40 pairs of computer-generated faces on a desktop computer, each pair of faces were either both happy (20 trials) or both sad (20 trials) faces. After looking at both faces next to each other on the screen, the child was asked to rate which face expressed the given emotion more intensely, with no time limit to make their choice. This task has shown to be reliable and valid in youth [51] and in early childhood aged 3–6 years [28]. The emotion recognition variables were operationalized as the raw value of correct recognition, with higher correct scores highlighting greater emotion recognition ability. Two variables were created; one variable used happy face trial scores that measured positive emotion recognition ability and another variable used sad face trial scores that measures negative emotion recognition ability.

Adverse Childhood Experiences

Adverse Childhood Experiences (ACEs) were measured using a sum composite score of a number of ACEs variables collected during the preschool-aged assessments (Time 1, Time 2, and Time 3) of the PDS study. Variables included various life events (see Table 1), from the life events section of the Preschool Age Psychiatric Assessment (PAPA) [52] or Child and Adolescent Psychiatry Assessment (CAPA) [53], parental psychopathologic variables (e.g., parent suicide, parent substance abuse, and any other parent disorder) from the Family Interview for Genetic Studies [54], and exposure to poverty based on family income to needs ratio at each assessment wave. Poverty was defined as an income to needs ratio less than 1, in accordance with federal guidelines. All variables were coded as absent or present (0 vs 1) and adverse experiences (other than poverty) were only counted multiple times if they were non-redundant (e.g. experience of abuse at one time point and then again at another time point). ACEs variables were standardized (z-scored) at each assessment wave, and then ACEs scores for preschool (ages 3–5.11) were calculated by computing means of the ACEs variables from assessments during that age range [55]. Variables included in the ACEs Score are listed in Table 1. As shown in Table 2, there was a wide range (0–7.33) of ACEs experienced in the current sample. Our mean of approximately 2.13 ACES is similar compared with other samples studying ACEs in early childhood (i.e., Mean 2.69 ACES [56]; Mean 1.1. ACES [42]; a majority of preschool-aged children experienced at least one ACE [57]).

Table 1.

Variables included in the ACEs Score

Poverty (Income-to-needs ratio <1)

Traumatic life events

Parent arrest

Parent hospitalization

Crash with motor vehicle, plane, or boat

Accidental burning, poisoning, or drowning

Attacked by an animal

Death of adult loved one

Death of sibling or peer

Domestic violence

Hospitalized, visited emergency department, or had invasive medical procedure

Man-made disaster

Natural disaster

Physical abuse

Sexual abuse, sexual assault, or rape

Witnessed someone threatened with harm, seriously injured, or killed  Physical violence or event causing death or severe harm

Other traumatic life event

Parental psychiatric disorders

Parental suicidality

Parental substance use disorder

Other parental psychiatric disorder

Note. Scored based on occurrence (i.e. each experience of an event was counted once unless the event occurred on multiple, separate, distinct occasions)

Table 2.

Demographic and Clinical Characteristics (N=251)

	Mean (SD) (unless noted otherwise)	Minimum	Maximum
Age,	5.43(.85)	3.13	6.99
Gender Female N (%)	120(47.4%)	--	--
Externalizing Disorder Diagnosis N (%)	115(46.4%)	--	--
Preschool Depressive Symptoms	2.33(1.50)	0	8
Sad Emotion Recognition	6.20(2.68)	0	12
Happy Emotion Recognition	5.07(2.63)	0	12
ACEs	2.13(1.52)	0	7.33

Note. Externalizing Disorder Diagnosis= diagnoses of ADHD, ODD, or CD; Preschool Depressive Symptoms= average of depressive symptoms across baseline sessions; Sad Emotion Recognition= number of correct responses for sad faces; Happy Emotion Recognition= number of correct responses for happy faces; ACEs= average of adverse child experiences across baseline sessions

Psychiatric Disorders and Symptoms

The Preschool Age Psychiatric Assessment (PAPA) [52, 58] is a validated, parent interview for diagnosing psychiatric disorders in preschoolers, aged 2 through 5. Masters level clinicians administered the PAPA to parents and dimensional scores of depressive symptoms were calculated across the preschool age assessments (Time 1, 2, and 3). Dimensional scores were created using an average of the core nine symptoms of major depressive disorder across the three preschool assessment waves, as we have done previously [59]. The PAPA was also used to calculate externalizing disorder diagnoses in children. Diagnoses of attention deficit hyperactivity disorder (ADHD), oppositional defiant disorder (ODD) and conduct disorder (CD) across the three preschool assessments were included (yes/no).

Statistical analyses.

Analyses were conducted using SPSS version 21. Associations with demographics and main variables of interest were completed using Pearson correlations and t-tests. In order to test the interactive effects of emotion recognition and adverse childhood experiences, we used a regression analysis, predicting depressive symptom severity across the preschool period. We completed two linear regressions using the PROCESS Macro [60] including mean centered variables of adverse childhood experiences (ACEs) and Emotion Recognition (EmoDiff) and their interaction. Specifically, the first regression examined moderating effects of negative (sad) emotion recognition while the second regression examined effects of positive (happy) emotion recognition. Analyses controlled for sex, mean preschool age, and externalizing diagnoses (ADHD, ODD, CD) collected during PAPA during the preschool-aged period. Given the substantial comorbidity between childhood depression and externalizing presentations [61], we controlled for externalizing diagnoses in order to assess the independent effects of ACEs and emotion differentiation on depressive outcomes given strong relationships of emotion differentiation with depression. Follow-up simple slopes probed significant interactions, examining the effects of low, moderate, and high emotion recognition scores on ACES in relation to average depressive symptoms during the preschool years and Johnson-Neyman tests were completed to examine when ACES and emotion differentiation were statistically associated with depressive symptoms. Cohen’s f² was calculated for effect sizes using R² from the interaction model and a second model with all variables except the interaction (.02=small, .15= medium and .35= large effect sizes). Data syntax materials can be found at: https://osf.io/b6wtp/.

Results

Sample characteristics are found in Table 2 and correlations among study variables in Table 3. Examining the influence of preschool ACEs and negative (sad) emotion recognition on current preschool depressive symptoms, when controlling for demographic (sex and mean preschool age) and externalizing disorder diagnoses, we found the model was significant F(6,244) = 13.07, R²= .24, p<.001 with a significant interaction between ACEs and emotion recognition of negative (sad) emotions (B(SE)= −.07 (.03), t=−2.08, p=.038, 95% CI: −.14, −.004) with a small effect size, Cohen’s f²= .04. Simple slopes analyses revealed that the influence of emotion recognition on ACEs was significant at scores of 1 standard deviation (SD) below the mean (B(SE)=..49 (.13), t=3.70, p=.0003) and at the mean (B(SE)=.29 (.10), t=2.90, p=.004) negative (sad) emotion recognition, but not at scores 1 SD above the mean of negative (sad) emotion recognition (B(SE)=.10(.14), t=0.68, p=.49). Stated otherwise, in the context of elevated ACEs, low and average ability to recognize differing intensity levels of negative (sad) emotion was associated with elevated current depressive symptom severity (Figure 1a). A subsequent Johnson-Neyman test demonstrated that ACEs and depressive symptoms was significantly associated when sad emotion recognition was 1.04 standard deviation above the mean, indicating that if a child is able to identify greater than 7 (1 SD above the mean) sad emotional faces, the relationship between ACEs and depressive symptoms was no longer significant. Additionally, as the number of sad emotional faces recognized decreases from 7, the relationship between ACEs and depressive symptoms becomes more positive. Specifically, when the lowest number of negative emotions are identified (0), B(SE)=.74(.23), t=3.22, p=.014), every additional ACE is associated with a .74 increase in depressive symptoms. When examining the role of positive (happy) emotion recognition, the model was significant, F(6,244) = 11.82, R²= .23, p<.001, but the interaction of ACEs and positive (happy) emotion recognition was not significant, (B(SE)=−.01 (.04), t=−.33, p=.74, 95% CI: −.09, .06) with an effect size of Cohen’s f² = 0 and so we did not follow up this analysis with simple slopes (Figure 1b).

Table 3.

Descriptive correlations among study variables

	Sad Emotion Recognition	Happy Emotion Recognition	Emotion Recognition Total	Preschool Depressive Symptoms	ACES
Sad Emotion Recognition	1	.354**	.825**	−.135*	−.097
Happy Emotion Recognition		1	.820**	−.062	−.009
Emotion Recognition Total			1	−.120	−.064
Preschool Depressive Symptoms				1	.292**

Note. Sad Emotion Recognition= number of correct responses for sad faces; Happy Emotion Recognition= number of correct responses for happy faces; Emotion Recognition Total= total number of correct responses across positive and negative emotions; Preschool Depressive Symptoms= average of depressive symptoms over baseline sessions; ACEs= adverse child experiences; averaged across three baseline sessions

^*p<.05;

^**p<.01 (2-tailed).

Figure 1.

Interaction of ACEs with Negative Emotion Recognition (A) and Positive Emotion Recognition (B) Predicting Concurrent Preschool Depressive Symptoms.

Because interaction tests are concurrent and as such we are unable to test temporal precedence, we completed post-hoc secondary analyses to also test a similar probable relationship: whether depressive symptoms interacted with ACES to predict a worse ability to recognize intensity of emotions. As such, we tested the moderating role of ACEs between preschool depressive symptoms and emotion recognition. The interaction between depressive symptoms and ACEs was not significant for negative (sad) emotion recognition (B(SE)= −.06(.11), t=−.54, p=.59) or positive (happy) emotion recognition (B(SE)= .01(.11), t=.12, p=.90), demonstrating some specificity to the ACEs and negative (sad) emotion recognition interaction predicting depressive symptoms.

Discussion

The current study considered whether ACEs are associated concurrently with depressive symptoms, and if recognition of emotional intensity, an early form of emotion differentiation, influences this relationship in preschool-aged children; building on extant literature in adult and adolescent samples. Supporting our hypothesis, low and moderate sad emotion recognition (but not happy emotion recognition) was associated with depressive symptoms in the context of elevated ACEs in preschool-aged children. Stated otherwise, findings revealed that in the context of adverse childhood experiences, preschoolers with poor and moderate ability to recognize sad emotional intensity exhibited elevated depressive symptoms. This study provides initial support for the importance of early-emerging negative (sad) emotion recognition on mental health outcomes starting as young as the preschool period. Specifically, when no negative emotions were identified in the current sample, each additional ACE experienced was associated with almost an entire additional depressive symptom. Given many children are faced with multiple ACEs in their lives, this could lead to high levels of depressive symptoms at young ages. On the other side, analyses demonstrated that if children are able to identify 8 negative emotions in this task, the relationship between ACES and depressive symptoms was no longer significant. This directly points to giving kids the tools to recognize the differing intensity of their negative (sad) emotions effectively in order to help them adaptively cope and decrease risk for depression, especially for kids facing serious adversity (i.e., multiple ACEs).

Similar to the adult literature [11], we found that lower sad emotion recognition ability is associated with greater depressive symptoms in preschoolers. Additionally, the current study also replicates previous findings in adolescents demonstrating negative emotion differentiation as a moderator between stressful life experiences (in our case ACEs) and depression [34]. Similar to Starr and colleagues (2019) [34] conclusions, we can consider our results in the context of a developmental diathesis-stress model of depression [62]. The diathesis stress model states that genetic, biological or cognitive factors interacting with environmental stress can result in disorders or other poor mental health outcomes [63]. Our findings are consistent with the model that lower emotion recognition ability can be considered the diathesis and when stressful life experiences (i.e., ACEs) are occurring in tandem, the result is increased depressive symptoms. As such, future research should consider the diathesis-stress model in emerging emotion differentiation or recognition of emotional intensity and depression relationships in preschool-aged children.

The role of sad emotion recognition in relation to depression has not previously been considered in children as young as preschool. The current study utilized a developmentally appropriate assessment of early-emerging emotion differentiation in young children, and demonstrated that this presentation of sad emotion differentiation (or recognition of sad emotional intensity) is significantly associated with depression, even at such an early stage of cognitive and emotional development. It also highlights that in the context of difficult psychosocial circumstances (i.e. ACEs), the ability to discriminate between varying intensity of sad emotions can potentially be protective against depressive symptoms. Specifically, when no sad emotions were identified in the current sample, each additional ACE experienced was associated with almost an entire additional depressive symptom. Given many children are faced with multiple ACEs in their lives, this could lead to high levels of depressive symptoms at young ages. On the flip side, analyses demonstrated that if children were able to identify 8 sad emotions, the relationship between ACES and depressive symptoms was no longer significant. This directly points to giving kids the tools to recognize the differing intensity of their negative (sad) emotions effectively in order to help them adaptively cope and decrease risk for depression, especially for kids facing serious adversity (i.e., multiple ACEs). Therapeutic treatments or school programs teaching kids how to recognize negative (sad) emotional intensity could help ameliorate or dampen the effects of having to endure serious negative events, such as living in poverty, witnessing domestic violence, or experiencing a natural disaster. Being able to intervene in early childhood by teaching negative emotion recognition could facilitate beneficial mental health outcomes that could potentially have long-lasting effects across development and the lifespan.

Also replicating the mixed emotion differentiation literature in adults [10–12], findings indicated null relations between positive (happy) emotion recognition and depressive symptoms. The non-significant findings between happy emotion recognition and depressive symptoms may highlight that even at this young age, emotion valence specificity is important. This could indicate that the relationship between effective emotional recognition with depressive symptoms is more acutely focused on primarily negatively valanced emotions beginning in early childhood. This may indicate that a narrower focus on distinguishing negative emotion intensity in early interventions for those facing adversity would be a more effective targeted strategy.

Although the current study was able to extend new findings to very young developmental samples, there were some limitations. First, the sample was enriched for preschoolers with early-onset depressive symptoms, meaning we did not have a representative sample of the general population and as such, results may not be generalizable to community samples. However, the number of ACEs experienced in our sample is comparable to other studies number of ACEs in a similar age group. Specifically, our sample demonstrated a Mean (SD) of 2.13(1.52) ACES, while a comparable preschool study examining the effects of ACEs in high stress, low socioeconomic status and racial minority single mothers reported 2.69(1.81) [56]. Additionally, general population samples demonstrate that the majority (56%) of preschool age children experience at least one ACE [57], paralleled by another study reporting 65.5% of children experienced at least one form of ACEs between birth to age 8 [64]. Taken together, our sample exhibited slightly higher number of ACEs than general preschool populations and similar mean ACES to samples recruited for high stress environments, demonstrating some generalizability to populations living in hardship, who are also prone to depressive symptoms. In addition, all measures were collected concurrently. Thus, we are unable to determine temporal precedence and nor can we make causal inferences. Future research should consider how sad emotion recognition ability in early childhood could prospectively influence depressive symptoms across development. Third, the measure of emotion recognition used in the current study differs from standard emotion differentiation assessments in adolescents and adults, which use daily sampling of emotional experience and examine variability in negative or positive emotional experience [3, 4, 9]. Although a purported developmental precursor to the emerging construct of ED, our measure may tap more into a basic emotion identification versus the fully-developed construct of differentiation evidenced in older populations. Future research may wish to use parallel daily sampling techniques to quantify emotion recognition in preschoolers and longitudinally assess whether emotional intensity recognition ability predicts emotion differentiation ability in developmental samples. Lastly, we are unable to compare scores from the emotion recognition task in our study to those of similar age groups because the modified, developmentally appropriate version of this measure has only been used in this sample (though it has been published in other papers using the same sample). Future research should aim towards creating preschool developmental norms of this adapted measure of emotion recognition to increase generalizability.

Implications

Recognition of emotional intensity is an important form of emotional processing at very early ages. In particular, in the context of adverse backgrounds and hardship, the ability to recognize negative (sad) emotional intensity could be a potential protective factor in relation to early childhood depressive symptoms. Knowing that adverse childhood experiences are common in communities faced by poverty and systemic bias, clinicians, teachers or parents working with children who face these early hardships may consider focusing on increasing negative emotion recognition in therapy. Helping children to label and identify sad emotions in others provides general socio-emotional development as well as the potential to buffer the effects that ACEs can have on the development of childhood depressive symptoms. This type of intervention could be important in both individual therapy and well as school-based programing.

Summary

ACES have repeatedly demonstrated associations with depression throughout childhood [45]. Conversely, good emotion differentiation and emotion recognition have demonstrated protective effects against psychopathology. The current study demonstrates that in the context of elevated ACEs, preschoolers with poor and moderate sad emotion recognition ability, a purported developmental precursor to emotion differentiation, exhibited elevated depressive symptoms. Teaching young children to recognize intensity levels of negative emotion may have significant positive effects on their ability to adapt to stressful experiences or environments.

Table 4.

Regressions predicting preschool depressive symptoms using sad and happy emotional intensity interacting with ACES

	Sad Emotion Recognition			Happy Emotion Recognition
Predictor	B(SE)	t	p	B(SE)	t	p
Constant	2.08(.64)	3.27	.001	2.11(.64)	3.28	.001
Age	−.01(.10)	−.13	.90	−.02(.10)	−.18	.86
Sex	−.15(.17)	−.84	.40	−.13(.18)	−.76	.45
Externalizing disorder	1.13(.18)	6.22	<.001	1.12(.18)	6.10	<.001
Emotion Recognition	−.05(.03)	−1.69	.09	−.04(.03)	−1.19	.24
ACES	.29(.10)	2.90	.004	.32(.10)	3.18	.006
Emotion Recognition x ACES	−.07(.03)	−2.08	.03	−.02(.04)	−.40	.69

Note: Preschool Depressive Symptoms= average of depressive symptoms over baseline sessions; Sad Emotion Recognition= number of correct responses for sad faces; Happy Emotion Recognition= number of correct responses for happy faces; ACEs= average of adverse child experiences across baseline sessions.