Emotional processing deficits in Italian children with Disruptive Behavior Disorder: The role of callous unemotional traits

Highlights

• We recruited 35 children with Disruptive Behavior Disorder, and 23 healthy controls.

• We used eye-tracking technology to evaluate eye gaze while the children processed emotional faces.

• High levels of CU traits were associated with lower sadness recognition.

• CU traits were related with a lower number and length of fixation on the eyes of sad faces.

• The length of fixations to the eye-region mediated sadness recognition.

Abstract

Research suggests that callous unemotional (CU) traits are associated with poor emotion recognition due to impairments in attention to relevant emotional cues. To further investigate the mechanisms that underlie CU traits, this study focused on the relationship between levels of CU and children's attention to, and recognition of, facial emotions. Participants were 7- to 10-year-old Italian boys, 35 with a diagnosis of Disruptive Behavior Disorder (age: M = 8.93, SD = 1.35), and 23 healthy male controls (age: M = 8.86, SD = 1.35). Children viewed standardized emotional faces (happiness, sadness, fear, disgust, anger, and neutral) while eye-tracking technology was used to evaluate scan paths for each area of interest (eyes, face, mouth), and for each emotion. CU traits were assessed using parent and teacher ratings on the Antisocial Process Screening Device. In the whole sample, elevated levels of CU traits were associated with a lower ability to recognize sadness, a lower number of fixations, and a lower average length of each fixation, specifically to the eye area of sad faces. In children with Disruptive Behavior Disorder diagnoses, high levels of CU traits were associated with lower duration of fixations to the eye-region on the eye area of sad faces, which in turns predicted lower levels of sadness recognition. The findings confirm that poor emotion recognition is associated with impairments in attention to critical information about other people's emotions. The clinical implications are discussed.

Introduction

Disruptive Behavior Disorders (DBDs), including Oppositional Defiant Disorder (ODD) and Conduct Disorder (CD), are common child mental health problems that are characterized by antisocial, hostile and aggressive behaviors and deficits in emotional regulation. In their most severe form, conduct problems involve the violation of the rights of others or the violation of major societal norms (American Psychiatric Association, 2013). The presence of Callous Unemotional (CU) traits in children with DBDs defines a subclass of children with higher etiological risk and poorer responses to interventions (for a review see Frick, Ray, Thornton, & Kahn, 2014). The main features of CU traits are a lack of empathy and guilt, shallow or deficient emotions, and a lack of care or concern about performance on tasks and other people's feelings (Frick & Ray, 2015). Recent research has suggested that while DBD children with elevated CU traits do less well in current interventions, they are not “untreatable” and can improve with intensive interventions when they are carefully tailored to their unique psychopathological characteristics (Hawes, Price, & Dadds, 2014). As such, research is needed to identify mechanisms for intervention that underlie the development of CU traits and can be targeted in early intervention. A better understanding of the mechanisms that promote and protect against high CU traits in youths, such as emotion processing, could inform treatments that directly target these mechanisms in DBD children.

There is substantial evidence that individuals with disruptive behaviors have difficulties in facial emotion recognition (Fairchild, Van Goozen, Calder, Stollery, & Goodyer, 2009; Short, Sonuga-Barke, Adams, & Fairchild, 2016; Sully, Sonuga-Barke, & Fairchild, 2015). Recently, impaired facial emotion recognition has been associated with elevated levels of CU traits in youths with disruptive behaviors (Blair, Leibenluft, & Pine, 2014). In brief, studies that investigated whether a child's ability to identify facial expressions accurately is related to CU traits show that the latter are associated with impairments in identifying expressions of emotion, especially distress emotions of sadness and fear (Blair & Coles, 2000; Blair, Colledge, Murray, & Mitchell, 2001; Dadds, El Masry, Wimalaweera, & Guastella, 2008). However, in a meta-analysis, Dawel, O'Kearney, McKone, and Palermo (2012) indicated that the direct or moderated effect of CU traits on emotion recognition is not limited to fear or sadness, and is evident to some extent for most types of emotions. Furthermore, not all studies have found an association between CU traits and emotion recognition impairments in clinical populations (Martin-Key, Graf, Adams, & Fairchild, 2018; Sully et al., 2015). It is important to note that CU traits appear to be associated with impairment in emotion recognition even in the absence of conduct problems. For example, Woodworth and Waschbusch (2008) found that children with higher levels of CU traits, regardless of whether they exhibit conduct problems, were less accurate in identifying sad facial expressions. Interestingly, in this study children with higher CU scores were more accurate in labeling fear than were children with lower CU scores. These contradictory findings are in line with the approach proposed by Dadds, Jambrak, Pasalich, Hawes and Brennan (2011); they suggested that children with CU traits are characterized by a more general impairment in attention to emotional stimuli, in this case, eye gaze to key features of the emotional face that may underlie a deficit in recognition across all emotions.

However, only two studies have investigated the relationship between levels of CU traits and attention to emotional stimuli, using eye-tracking systems to detect eye gaze impairment. Dadds et al. (2008) tested whether high levels of CU traits are associated with reduced attention to the eye region of other people's faces, in a sample of typically developing youths. Findings from Dadds' study (2008) showed that high levels of CU traits predicted poor fear recognition, a lower number and duration of eye fixations, and fewer first foci to the eye region. There were no differences in gaze indices to the mouth region. Dadds et al. (2008) also showed that the simple manipulation of asking youth to look at the eyes results in increased emotion recognition. These evidences suggest that the attention to eye region of others might mediate the relation between CU traits and emotion recognition. Recently, Martin-Key et al. (2018) found that CU traits predicted reduced fixations to the eyes, but this was verified only for surprised expressions.

At the neural processing level, a wealth of studies that suggested involvement of the amygdala in both attending to and interpreting facial emotion expressions and monitoring of the eye-gaze of other people (Haxby, Hoffman, & Gobbini, 2000). Thus it is instructive to note that fMRI investigations have revealed an association between CU traits and selective impairment in empathic responses characterized by reduced amygdala responses to fearful expressions (Jones, Laurens, Herba, Barker, & Viding, 2009; Marsh et al., 2008; White et al., 2012a). In particular, White et al. (2012b) demonstrated that the amygdala hypoactivity observed in response to fearful faces in young patients with DBD and high CU traits is not secondary to an attentional deficit, but is specifically related to the CU component of psychopathic traits.

No study has yet investigated the relationship between the levels of CU traits and impairment in emotion recognition and/or in eye movement behavior using eye-tracking in children with a DBD diagnosis. In the present study, we investigated the role of CU traits in determining emotional processing in children with a DBD diagnosis, using a remote eye-tracking system. We argue that investigating the association between levels of CU traits and emotional processing in a clinical sample is important for the development of new intervention models for DBD children. If a specific association between CU traits and emotional processing deficits is confirmed in a clinical sample, further development of interventions which focuses on improving children's abilities to process emotions might reduce children's CU traits (e.g. Dadds, Cauchi, Wimalaweera, Hawes, & Brennan, 2012; Hubble, Bowen, Moore, & Van Goozen, 2015).

We measured levels of CU traits and externalizing problems continuously across a sample of DBD and typically developing children, and tested in the whole sample whether the level of CU traits, externalizing behavioral problems, age, intelligence quotient, income and group membership are associated with emotion recognition deficits and impairment in eye gaze to key features of emotional faces. Starting from the evidence found in Dadds et al. (2008), we also explored a mediation model linking CU traits to emotion recognition deficits through attention to the eye-regions. Unlike Dadds et al. (2008) and Martin-Key et al. (2018) who focused their studies on adolescents, we chose to analyze the relationship between CU traits and the processing of emotionally salient stimuli in younger children. This may enable us to see whether the difficulties that were found in the aforementioned studies emerge in the developmental phases preceding adolescence at a time when early intervention may be more successful.