Recent work has shown that paranoia - excessive suspiciousness of others - exists on a spectrum of severity in the adult general population. Yet little is known about either the nature of mistrust in children or whether studying paranoia in children could increase our understanding of the aetiology of adult paranoia and inform early prevention strategies. The current thesis, comprised of three main studies, adopted a hitherto lacking developmental perspective to examine social mistrust in middle childhood. The first goal was to assess the structure, prevalence, correlates and short-term stability of childhood mistrust in nonclinical samples drawn from two different countries (the UK and Hong Kong). Classroom-based surveys of 8- to 14-year-olds from the UK (N = 1,086) and Hong Kong (N = 1,470) were carried out between 2011 and 2014. A new measure developed for the study was administered: The Social Mistrust Scale. Analysis of the UK data indicated three factors of mistrust with the scale, based upon the source of threat: home, school, and general mistrust. This factor structure was replicated in the Hong Kong sample. Consistent with the adult literature, the majority of children reported a few symptoms of mistrust and a few children reported many. Younger children reported higher levels of mistrust than older children. Overall, mistrust was unrelated to gender, verbal ability and country. In both countries, mistrustful children (those scoring 7 on the Social Mistrust Scale) were more anxious, had lower self- esteem, and were more aggressive and callous than trusting children. Being mistrustful in school in particular, above and beyond home and general mistrust, was associated with elevated levels of anxiety. A one-month test-retest study (N = 271) showed that mistrust was a stable construct (ICC = .80). The second goal was to examine children’s definitions and reasons for social trust and mis- trust. This was a large qualitative examination of interviews with children, in order to learn more about the phenomenon at this age and generate future research questions to test. Semi- structured interviews were conducted with 122 children classified as trusting (bottom-15%) and mistrustful (top-15%) on the Social Mistrust Scale (SMS) and matched on country, gender, age and school. Detailed thematic analysis of 95 randomly selected transcripts (80%) was con- ducted to identify broad themes surrounding trust as well as mistrust. In this study agreement between interviewer ratings and child-reported mistrust was also compared. It was found that the majority of children defined social trust as secret keeping and sharing, and viewed trust as important to relationship development and maintenance. Responses to questions about mistrust revealed three recurring sub-themes. The first was a link between mistrust, popularity and bul- lying. Children often spoke about distrusting popular children with their secrets and concerns because popular children generally had many friends or were bullies. Worries and anxieties about being targeted formed the second theme. For a small minority of children these feelings were evoked by ideas of being followed, spied on and attributing hostile intent to others. The third recurring theme pertained to how children coped with their worries and to whom they would go to for help (i.e., friends, family, teachers). A minority of children expressed persistent worries and suspiciousness of others. Absolute agreement between interview-assessed and self- reported mistrust was moderate (k = .46). This study generated several questions: (i) what is the relationship between mistrust and classroom social status? (ii) are mistrustful children more likely to experience worries and loneliness because they are bullied more often than trusting children? and (iii) to what extent are children’s suspiciousness unfounded? The final main goal was to test the association with childhood mistrust and a number of potential causal factors identified from the adult literature. Cognitive processes (i.e., reasoning bias, theory of mind and executive function) and psychosocial risk factors (i.e., bullying, lone- liness, peer-rated social status, and hostile attribution bias) were studied. 118 children from the follow-up sample completed a battery of tasks assessing social mistrust, theory of mind, execu- tive function, reasoning bias (the jumping to conclusions beads task), hostile attribution biases, bullying, school exclusion and depression. It was found that mistrust in the children was as- sociated with all psychosocial risk factors (depression, loneliness, bullying, victimisation) and peer-ratings of trustworthiness/popularity but not with parent-ratings of mistrust. Mistrustful children were more likely than trusting children to attribute hostile intent in others and to report persistent peer victimisation, controlling for children’s likelihood of being a bully, feelings of loneliness and depression. These groups did not differ in tests of theory of mind, executive function and cognitive reasoning bias performance, which tentatively suggests that social and affective processes rather than cognitive factors may play a role in the occurrence of childhood mistrust. ii Overall, this thesis presented evidence that: (i) Social mistrust is prevalent in a minority of children, and it is associated with both internalising and externalising problems; (ii) Qualitative interviews indicated that mistrust was often well-justified but that a minority of children may well be having excessive suspiciousness about being targeted; (iii) Mistrustful children (especially with mistrust about school) report persistent victimisation and hostile attribution bias but do not show biases in non-affective cognitive performance compared with trusting peers; and (iv) There is moderate agreement between self-report and interviewer assessments of paranoia, child and peer ratings of mistrust but not with parent ratings. This thesis began the task of researching a developmental perspective on childhood suspicious- ness, extending the work in adults. Mistrust is present in children and associated with symptoms of mental health problems and adverse experiences. The extent to which the fears were unfounded (i.e. true paranoia was assessed) was not established in the thesis nor the causal direction of the associations found. Continued research on social mistrust in community children and beyond may provide promising avenues to earlier preventions and better treatments of paranoia.