The data is drawn from studies conducted/published between 1990 and 2015 that investigated the relationship between interpretation bias and anxiety in children and adolescents. The data was originally gathered to conduct a meta-analysis investigating the presence and strength of the relationship between interpretation bias and anxiety in children and adolescents, as well as assess certain study characteristics as moderators. Relevant studies for inclusion in this data were identified via three channels: through data bases such as psychInfo using specially selected search terms, searches of reference lists of accepted papers and contacting authors for unpublished work. The abstracts, and subsequently full texts, of all studies identified through these channels were assessed for eligibility against a set of criteria that also assessed the studies for quality. The data consists of the effect sizes related to the relationship between interpretation bias and anxiety found in studies that fulfilled the eligibility criteria. Alongside each effect size are study characteristics such as demographics, measures utilized and design features are coded and were used as moderators in the meta-analysis. This is the first data-set of its kind to code and detail all available studies that assess the relationship between interpretation bias and anxiety in a young population in this way. Childhood anxiety affects daily functioning, such as school attendance, in around 6% of children. Untreated childhood anxiety disorders show associations with depression and suicidal ideation. Research examining anxiety development in children is therefore important for guiding early intervention and treatment with a view to preventing long-term mental health problems. Cognitive theories state that anxiety is underpinned by dysfunctional cognitive processes, anxious children show at least two types of cognitive bias: an attentional bias, where attention is disproportionately allocated to threatening stimuli, and an interpretation bias, which increases the likelihood of ambiguous stimuli being interpreted as threatening. However the exact relationship between these cognitive biases and anxiety development is unclear; biases may precede anxiety symptoms, acting as an early indicator of anxiety risk and/or increasing the likelihood of subsequent anxiety problems, or biases may be a symptom of anxiety but not play a role in the development of anxiety. In order to ascertain whether cognitive biases play a role in the onset of anxiety, research is required that tracks biases and levels of anxiety across childhood. Current understanding of how cognitive biases influence anxiety development is limited due to methodological problems in measuring cognitive biases. Cognitive biases have been assessed through subjective self-report or basic experimental tasks. These methods require adequate language skills and attentional capacity, hence are inappropriate for young children. Objective measures that do not rely on self-awareness, language or indirect reports may provide a more reliable measure of information processing in children. However, research using objective measures to assess information processing biases in children is scarce. This PhD aims to develop and validate objective methods of measuring cognitive biases in children aged 4-8 years, with a long-term aim of elucidating the relationship between bias and anxiety across development. A number of different methods will be used in a series of studies to tap attentional bias and interpretation bias respectively. Eye-tracking provides reliable information about visual attention and has been used within adult anxiety studies. We will use eye-tracking to investigate children's focus of attention whilst viewing faces expressing an emotion. Differences between the attentional allocation of children, who vary in anxiety level and age, can be assessed. ERP components such as the EPN and LPP may also give an indication of differing attentional resources allocated to specific expressions. Physiological regulation (i.e. heart rate variation) could be used to objectively measure children's stress/anxiety during the task providing additional insight into their processing of the stimulus. To investigate interpretation bias facial electromyography (EMG) will be used to establish a child's emotional reactions when asked to respond to the valence of ambiguous sounds. Facial EMG measures the electrical activity of facial muscles responsible for expression of emotion and indicate, for example, reaction to threat, thus indicating children's expectations and feelings about the stimuli. Initially measures will be developed using a community sample screened for high and low anxiety. A clinical sample of children will then be recruited to validate the final measures. It is expected that the high and low anxious participants will show differences in their processing of stimuli as assessed using the objective measures. Using a cross sectional design age-related changes in the relationship between biases and anxiety will be investigated.

Searches were conducted on various data-bases, including those containing theses, using search terms to find all relevant studies to the aim of the meta-analysis. Searches were conducted on 6th August 2015 and the search was restricted to studies published since 1990. Preliminary searches indicated no studies with the relevant age range were published prior to 1990. References of accepted studies were also searched for applicable papers, and first and corresponding authors of accepted papers were contacted to request unpublished data. 29 Authors provided unpublished data, 24 of which met criteria therefore were included in the meta-analysis.