The world of robotics, like that of all technology is changing rapidly (Melson, et al., 2009). As part of an inter-disciplinary project investigating the emergence of artificial culture in robot societies, this study set out to examine children’s perception of robots and interpretation of robot behaviour. This thesis is situated in an interdisciplinary field of human–robot interactions, drawing on research from the disciplines of sociology and psychology as well as the fields of engineering and ethics. The study was divided into four phases: phase one involved children from two primary schools drawing a picture and writing a story about their robot. In phase two, children observed e-puck robots interacting. Children were asked questions regarding the function and purpose of the robots’ actions. Phase three entailed data collection at a public event: Manchester Science Festival. Three activities at the festival: ‘XRay Art Under Your Skin’, ‘Swarm Robots’ and ‘Build-a-Bugbot’ formed the focus of this phase. In the first activity, children were asked to draw the components of a robot and were then asked questions about their drawings. During the second exercise, children’s comments were noted as they watched e-puck robot demonstrations. In the third exercise, children were shown images and asked whether these images were a robot or a ‘no-bot’. They were then prompted to provide explanations for their answers. Phase 4 of the research involved children identifying patterns of behaviour amongst e-pucks. This phase of the project was undertaken as a pilot for the ‘open science’ approach to research to be used by the wider project within which this PhD was nested. Consistent with existing literature, children endowed robots with animate and inanimate characteristics holding multiple understandings of robots simultaneously. The notion of control appeared to be important in children’s conception of animacy. The results indicated children’s perceptions of the location of the locus of control plays an important role in whether they view robots as autonomous agents or controllable entities. The ways in which children perceive robots and robot behaviour, in particular the ways in which children give meaning to robots and robot behaviour will potentially come to characterise a particular generation. Therefore, research should not only concentrate on the impact of these technologies on children but should focus on capturing children’s perceptions and viewpoints to better understand the impact of the changing technological world on the lives of children.

The resource, or strength, model of self-control (Baumeister, Heatherton & Tice, 1994) suggests that individuals possess a limited resource of strength, or energy, which is depleted by acts of self-control, leading to reduced performance on a subsequent, unrelated, self-control task (Muraven, Tice & Baumeister, 1998). This decrease in self-control has been labelled ‘ego depletion’ (Baumeister et al., 1998). Review of the research literature reveals an impressive array of effects linked with ego depletion; however, the majority of these studies have focused on depletion in adults. The present study aimed to extend the research literature in this area by investigating the ego depletion effect in 89 primary school children aged between 10 and 11 years. A dual-task procedure was used to investigate the potential ego depletion effect of a brief thought suppression task upon a subsequent task of receptive attention. In addition, the current study aimed to find out whether trait mindfulness (i.e., a disposition towards open and non-judgemental awareness of one’s self and attention to the moment) can counteract ego depletion in children (Baer, Smith, Hopkins, Krietemeyer & Toney, 2006; Bishop et al., 2004; Brown & Ryan, 2007a). No significant effect of ego depletion on performance in the second self-control task was found. A borderline significant effect of ego depletion on the children’s perceived difficulty of the second self-control task was found. Trait mindfulness was found to be a significant predictor of children’s perceived difficulty of the second self-control task. No significant moderation effect of ego depletion by trait mindfulness was found. Implications are explored and future directions discussed.

Social learning is one important way that children learn about the world. This thesis presents and discusses several current social learning theories, exploring how they explain different facets of social learning. In particular, I examined the naïve theory of rational action, the theory of natural pedagogy, the ideomotor approach to social and imitative learning, and the normative account of social learning. Each theory is reviewed on how it explains four facets of social learning: imitation, emulation, action understanding or interpretation, and the consideration of varying social and situational circumstances. The review shows that each theory focuses on only one or two facets, often providing very limited discussion (if any) of the others, and none of the theories systematically varies its predictions of a learner's behaviour as a factor of multiple social and situational circumstances. By means of six empirical studies I show that the social and situational circumstances strongly influence social learning, and that none of the discussed theories can account for the findings at large. I also argue that the current social learning theories explain developmental shifts in different biases that affect social learning rather than core mechanisms of social learning, and that what is needed is a strong social learning theory that explains multiple facets of social learning, in particular making differential predictions as a factor of varying social and situational circumstances.

This thesis focuses on the influence of younger siblings on firstborns’ understanding of minds in middle childhood. This topic was investigated in the context of the Cardiff Child Development Study (CCDS): a prospective longitudinal study of firstborn children and their families living in Wales, UK. As reported in Chapter 3, presence of a younger sibling resulted in an advantage on a second-order false belief task at 7 years of age. However, this advantage was only true for firstborns who experienced the birth of a sibling after their second birthday. It was hypothesised that the positive contribution of a younger sibling is mediated by changes to features of mother-firstborn conversation in dyadic interactions. To test this hypothesis, in Chapter 4 I described an expanded internal state language coding scheme for analysing mothers’ speech at 6 months, 21 months and 7 years. Mothers’ references to their 7-year-olds’ cognitive states positively predicted understanding of second- order false belief. As reported in Chapter 5, mothers who had a second child referred to cognitive states more than those with one child in middle childhood. Mothers’ variety of references to cognitive states partially mediated the association between presence of a sibling and firstborns’ understanding of second-order false belief. In Chapter 6, I investigated the families who experienced the birth of an early arrival younger sibling. Early arrival siblings were predicted by mothers’ symptoms of conduct disorder. Mothers who had an early arrival sibling present by 21 months referred to cognitive states less than mothers who did not. The findings in this thesis contribute to knowledge about the influence of younger siblings on the child’s development of theory of mind. The findings demonstrate the importance of examining how the arrival, and timing of arrival, of a younger sibling affects the dynamics of children’s other close relationships that influence children’s developing understanding of minds.

The work presented in this thesis comprises the execution and evaluation of a pilot quasi- Randomised Controlled Trial of Reciprocal Imitation Training (RIT) in an attempt to replicate previously reported effects of RIT on imitation skills in children with autism and an evaluation of the effect of RIT on brain functioning. Children with autism were randomised into two groups, Treatment and Wait-List Control, and were assessed before and after intervention. Behavioural measures of spontaneous and elicited imitation were used to assess change in imitation. Event-Related Potentials (ERP) and Electroencephalography (EEG) techniques were used to index changes in human action processing as well as global social and non-social processing. An increase in spontaneous, social imitation skills was evident in the Treatment group compared with the Wait-List Control group. Also, ERP measures assessing auditory human action processing reflected differences in processing at outcome between the Treatment and Wait-List Control group. However, no effect of RIT was observed on global social or non-social neural processing. Together, these findings make contributions towards evaluating the efficacy of RIT as an early intervention program for children with autism with evidence of an impact on social imitation skills at the behavioural and neural level.

The present study is concerned with the relation between children's self-acceptance, and their acceptance by their peers, as well as the developmental processes involved in the establishment of the self-image during middle childhood. 144 children divided into four age groups (age 7.5, 8.5, 9.5 and 10.5) were included in the study. Children's self-image was investigated in terms of three variables: 1) self-acceptance 2) uncertainty or inconsistency of self-evaluation, and 3) a social versus personal orientation in evaluating both self and peers. Social adjustment was studied in terms of four variables: 1) sociometric status, 2) reciprocal choices, 3) insight into peers' choices, and 4) peer interactions during play situations. In the first part of the study, the equivalence of three measures of self acceptance: 1) self-esteem derived from an interview technique, 2) self-acceptance derived from a Q sort technique, and real self-ideal self congruency (derived from Kelly's repertory grid technique) was borne out. Self-acceptance and real self-ideal self congruency were found to be positively associated, though not entirely overlapping. The second part of the study was concerned with some developmental trends, which emerge during middle childhood in the self-image, as well as in peer-interaction. In this respect it was found that: 1) self-acceptance linearly decreases with age. 2) real self-ideal self congruency decreases with age, though a quadratic rather than linear age trend component was found. 3) Children's uncertainty with regard to self-evaluation linearly decreases with age. 4) The amount of peer interaction in play situations linearly increases with age. Social self-orientation (in terms of the ideal self) was found to increase with age. A parallel personal selforientation linearly decreased with age. A differentiation according to a social versus personal orientation in evaluating peers and self was investigated by principal component analysis. In the fourth part of the study a curvilinear relation between self-acceptance and acceptance by peers, tested separately in each of the eight subgroups (divided according to age and sex), was confirmed. Furthermore, the significance of this relation linearly increased with age.

This study aims to assess whether children have measurably different relationships with some of the familiar adults in their everyday world. Assessment is made by observing preschool children in a conventional park playground, and evaluating whether their activities and social behaviour differ depending on the caretaker type/s accompanying them. The study is in three parts. The first is an empirical validation of the research venue. Since playgrounds are rarely used in developmental research, it is important to establish that the patterns of behaviour observed there are generalisable. The second part is a comparison of children accompanied by either mother, father, grandmother, nanny, or peer's mother. All 5 caretaker types were found to exert some unique influences, although similarities outweighed differences. The third part compares the effect of a single caretaker accompanying a child with the effect of two caretakers. This allows for assessments of the robustness of caretaker influences. Caretaker effects were found to be dynamic, with two caretakers having an effect which could not be predicted from their effects when alone with a child. There is also evidence of some caretakers having a more dominant effect on children's behaviour than others. It is concluded that caretaker roles provide consistent predictors of children's activity and social behaviour although roles are probably only one of a number of salient factors in determining the effect of different familiar adults on children.

About fifteen years ago, the unreflecting acceptance of such concepts as right and left as simple polarities, virtually prevented most serious researchers from investigating closely how a child comes to differentiate and name both sides of his body and succeeds in doing so on other persons as well as on himself. Work done by Spionek and Benton in particular has highlighted the various stages through which the development of the right-left concept progresses. Initially it is identified on oneself then 011 other people and on objects. The problem now is to isolate the variables which have a determining influence; this involves an evaluation of the extent of their influence and their interaction. A sample of eighty boys and girls, four to twelve years old was selected and seven tests covering the variables thought to be relevant were administered to them, namely: body schema, vocabulary, abstract reasoning, space relations, handedness, and motor coordination; a measure of knowledge of right and left was also employed. The data obtained on those tests together with the age element were submitted to Principal Component Analysis. This statistical analysis shows that maturation is the most Important of all components extracted. A second component was one of directional orientation and a third, of gross visuo-motor coordination. A fourth factor also emerged involving handedness and right-left discrimination on others. This investigation ofhandedness gave rather puzzling results; further examination suggested that the concept was oversimplified, it is an intricate multidimensional characteristic, rather than a unitary one. It is concluded that verbal ability plays a part throughout the whole evolution of the right-left concept while age and body image are relevant only when the child is applying the concept to himself; space orientation is correlated with this application on other persons.

Theoretical perspectives on risk and resilience are important in understanding anxiety in children and adolescents, and the development of intervention approaches. A systematic review of the literature was conducted using a manual search and electronic databases PsychInfo, Medline, Embase and Web of Science. Key research studies evaluated universal school-based interventions aimed at reducing anxiety and depression, including those with active, passive or no control groups, and with participants aged 3 – 17 years old. The results indicated that universal interventions, delivered by teachers or mental health professionals, can be effective in reducing anxiety, with a small mean effect size (d = 0.23). Most programmes were brief (i.e., 8 - 10 weeks) and targeted children > 9 years old. Implications for future research include the inclusion of younger populations, multiple informants for outcome measures, and a theory based approach to understanding the factors associated with positive change. The empirical study explored the impact of a universal cognitive behavioural programme for young children in reducing anxiety (and associated behaviours), improving attention and peer relationships. The study also looked at whether attentional control (AC) was important in understanding change. Sixty children (aged 4 – 5 years) attending a mainstream school received the 12-week intervention. Children completed a computer task to assess AC and completed a sociometric status measure before and after the intervention. Teachers completed measures to assess children’s anxiety and peer relationships. The results showed that anxiety reduced (p <.05), and peer problems and pro-social behaviour improved (p < .001) following the intervention. AC also improved (p <.05), and this change was associated with an increased number of reciprocal friendships. The findings add to the literature on the effectiveness of universal interventions in reducing internalising difficulties, and offer some insight into the factors involved in understanding positive outcomes.

Ce travail raconte un parcours professionnel guidé par la pratique, la curiosité et l’envie de comprendre. Il propose un modèle thérapeutique pour prendre en compte plus finement la complexité des situations cliniques en tricotant dans un ensemble congruent les informations relationnelles de l’ici et maintenant avec celles de l’histoire personnelle de chacun des acteurs de la relation. Une présentation de différentes publications illustrent ce cheminement de pensée à travers des expériences diverses pour arriver à la modélisation d’une approche thérapeutique de l’individu en relation / This work is about a professionnal journey guided by clinical practice, curiosity and desire to understand. It proposes a therapeutic model to better apprehend the complexity of clinical situations by interweaving in a congruent fashion relationnal information from the here and now with the historical personal information of every individual involved. A presentation of different publications highlights this evolution of thinking as expressed in different experiences which provides a therapeutic approach model of the individual in relation with others