Downloads & Free Reading Options - Results

Adolescence is a period during which many mental health difficulties first arise (Kessler, Berglund, Demler, Jin, Merikangas, & Walters, 2005). Difficulties in emotion processing have been associated with an increased risk of mental health difficulties in adolescence, including internalising difficulties such as anxiety and depression (Garnefski, Kraaij, & van Etten, 2005; Silk, Steinberg, & Morris., 2003). Internalising symptoms in adolescence, even at subclinical levels, can persist into adulthood (Petersen et al., 2018; Winefield, Hammarström, Nygren, & Hägglöf, 2013). Understanding early deficits in emotional processing is therefore essential for developing our understanding of how and when internalising problems first arise. There is a wealth of previous research that has investigated the presence and nature of emotion processing deficits in children and adolescents with internalising problems. Children and adolescents with anxiety disorders perform significantly less accurately on facial emotion recognition tasks compared to non-anxious controls (Collin, Bindra, Raju, Gillberg, & Minnis, 2013). Equally, a general deficit in emotion recognition in depressed individuals has been indicated in reviews and meta-analyses of studies involving both adults and children (Kohler, Hoffman, Eastman, Healey, & Moberg, 2011; Dalili, Penton-Voak, Harmer, & Munafo, 2015). When looking specifically at how anxious and depressed youths process different facial emotions, the literature indicates that it may be more nuanced than a general deficit across the board. For example, adolescents with depression have demonstrated a dual deficit in emotion processing whereby they are more accurate than non-depressed controls at recognising sad faces and less accurate at recognising happy faces (Auerbach, Stewart, Stanton, Mueller & Pizzagalli, 2015). However, the recent literature exploring how well children and adolescents with internalising problems recognise different emotions has yielded mixed results. For example, some studies have found that children and adolescents with increased internalising symptoms are more accurate at identifying fearful faces (Simcock et al, 2020) whilst other studies have found the opposite pattern, with anxious adolescents performing worse than healthy controls when identifying fearful expressions (Wieckowski et al, 2016; Dede, Delk, & White, 2021). Despite these discrepancies, research has consistently identified an attentional bias towards threatening or fearful faces in those with greater internalising symptoms (see Valadez, Pine, Fox, & Bar-Haim, 2022 for review). Collin and colleagues’ (2013) review found that anxious youth demonstrated an attentional bias towards angry or fearful faces relative to controls (McClure et al., 2007; Monk et al., 2006; Waters, Mogg, Bradley, & Pine, 2008), whilst children with both BPAD and anxiety were found to show a bias towards threatening faces (Brotman et al, 2007). Children and adolescents with BPAD were also found to be more afraid in response to neutral faces than healthy controls, further suggesting a bias towards perceived threat (Brotman et al, 2010; Rich et al, 2006). This attentional bias towards threatening faces is likely to impact these children and adolescents’ abilities to recognise threatening emotions, such as fear. The relationship between emotion recognition and internalising problems should be explored in more detail to shed light on different findings found thus far. In order to study developmental processes, we must take a developmental perspective to conducting research. Previous studies have therefore cited the need for more longitudinal research. Longitudinal analyses are required to determine whether behavioural and neurobiological differences in emotion processing can in fact be understood as early indicators of later psychopathology or whether they might be ‘transient developmental variations’ (Blok et al, 2021). This may depend on the impact of biological, psychological or social risk and protective factors, which can also be investigated using longitudinal data. Findings from this kind of research could have important implications for mental health prevention and early intervention planning. This study’s main objective is to better understand the relationship between the processing of fearful faces and internalising traits in adolescents. Specifically, I aim to better understand how individual differences in internalising traits are associated with the recognition of fearful faces as well as the underlying associated brain activity. I will do this by utilising an existing dataset – the Adolescent Brain Cognitive Development (ABCD) study. In the first (cross-sectional) part of my study, I will use data from the most recent ABCD data collection wave, which is from individuals aged 11-12 years. I will consider both performance on a facial emotion recognition task (the Emotional N-back task; Casey et al, 2018) in response to fearful faces and brain activation in key regions of interest during this task in order to provide a thorough exploration of this association. Key brain regions that are thought to underpin the processing of fearful faces include the amygdala, the anterior cingulate cortex (ACC), the ventromedial prefrontal cortex (vmPFC) and the ventrolateral prefrontal cortex (vlPFC; Del Piero, Saxbe, & Margolin, 2016;). Hyperactivation in the amygdala (Hulvershorn, Cullen, & Anandl, 2011; Yang et al, 2010; Monk et al, 2008) and ACC (McClure et al, 2007; Gotlib et al, 2005) have been associated with increased internalising difficulties, whilst hypoactivation in the vmPFC (Pitskel et al, 2011; Johnstone et al, 2007) and vlPFC (Pavuluri et al, 2007; Passarotti et al, 2010) have been associated with increased internalising difficulties. In the second (longitudinal) part of my study, I will explore whether earlier facial emotion recognition ability and the associated underlying brain activity can predict later internalising symptom severity. I will explore this by using ABCD data obtained from two time points: time 1 (age 9-10 years) and time 2 (age 11-12 years). Beyond this, I will also explore whether any key developmental or social variables might also predict internalising symptom severity in early adolescence, as this may highlight which young people are most vulnerable. Specifically, this includes sex, pubertal timing, parental mental health, and socioeconomic status. These factors have all been identified as important risk factors for developing internalising difficulties (Lynch et al, 2021; Reiss, 2013). This project will be published as a doctoral student thesis. It forms part of a larger project exploring predictors of externalising and internalising within the ABCD cohort.