The thriving of aggressive behaviour and delinquency in adolescence is a worrying subject for society in general. Analysing and knowing the determinant factors is the main objective in the research focused on prevention and intervention. The bibliography review carried out in the last decade highlights the psychological and social variables related to this problem. The empirical research selected emphasises a series of psychological and behavioural variables as well as social variables related to delinquent and aggressive behaviours. Among the first, the following stand out: 1. Those which are a protection factor: cognitive processes (moral reasoning, perspective taking or the ability to put oneself in the place of another), prosocial behaviour, coping mechanisms that are productive and efficient when facing situations that provoke tension or require a solution to a problem, emotional intelligence, emotional self-regulation (especially to control anger), positive emotions (empathy in its affective component of concern about others stands out). 2. Those which are risk factors of delinquency and aggressive behaviour: temperament (in particular impulsiveness, main predictor of aggressive behaviour), depressive symptoms, negative emotions (anger stands out), coping mechanisms that are unproductive or inefficient when facing problems or conflict, aggressive behaviour, substance abuse, behavioural problems. Among the social variables stand out: abuse sequences and those related to the socialisation process: family, school and peers. The parenting styles are considered risk factors if permissiveness dominates the relationship with the parents, the scarce involvement in parenting or negative control (negative evaluation of the children, punishments, excessive rules and criticism), or on the contrary they are protection factors if it is affection, communication and rules, that dominate. In the relationship with peers, the secure attachment or quite the opposite, the relationship with rebel peers or gangs are related to a more or less adapted behaviour in the home or the school environment. Furthermore, research in this area analyses the processes involved from gender perspective and age. In the last decade, an increase of violence and family aggressions, in particular between adolescents and their parents has been verified. There have been given different definitions of this kind of violence, but in general, the child to parent violence is that in which the child acts consciously with the intent to obtain or maintain power, control and dominance over their parents in a repeated manner, hurting their victim, with the immediate end to obtain what they desire causing psychological, financial or physical violence. The child to parent violence presents a characteristic violence cycle, with lack versus excess of parental limits, where the modus operandi in the aggressor-victim pair becomes a coercive cycle of submission-hostility / hostility-submission, where the abused parents notice that their educational resources are not effective. In fact, when they use these resources (reprimands or punishments), the child responds by increasing, both in intensity and frequency, their violent behaviours. Due to all this, the relationship between parents and children finds itself trapped in a process of action-reaction, where parental submission used to achieve peace in the home environment, causes an increase of the demands of the child against the expectations of the victim (Aroca, 2013). At the beginning of this decade the Attorney General’s Office Publication 1/2010 highlighted that “… the domestic abuse carried out by minors is spreading lately in a way that it should, at least, be qualified as worrying.” The main causes are due to “a permissive society that educates about the rights of the children, but not about their responsibilities, where the motto “without limits” has wrongly transpired (...) preventing a correct maturation process. There are parents who not only are not able to demand respect, but that diminish the authority of the teacher, the Police or other citizens when, in defence of family life, reprimand their descendants.” The data shows an increase in recent years. Attorney General’s Office 2015 Yearly Report shows that the the numbers of family violence, unfortunately, remain stable: 4,753 cases open in 2014, compared to 4,659 in 2013 and the 4,936 in 2012. The profile of the aggressor is that of an adolescent between 14 and 16 years of age, although recent research widens the aggressors age range to between 10 and 18 years of age, more frequently male, middle and upper class, who is not used to having his desires denied and with a low frustration threshold. This profile highlights the need to deepen the analysis of the psychological and social variables that influence violent behaviour against parents. Another look at the latest published research in the area indicates: 1. A higher occurrence in the age range between 12 and 16 years of age. 2. The importance of the parenting styles and the family environment: family cohesion, affection, and communication act as a protection factor and are considered positive family contexts, and on the contrary, family conflict, parental negligence to attend the needs of their children and the little involvement in their upbringing are risk factors for the child to parent violence. 3. As for the personal and social variables, drug use, school failure, exposure to violent situations and environments, depressive symptoms, difficulty to express feelings, lack of empathy, few prosocial behaviours, behavioural problems in and out of the home and anti-social behaviour stand out as facilitators of child to parent violence. 4. The kind of child to parent violence is characterised by more physical violence from boys towards their parents and more verbal violence from girls towards their parents. The mother appears more frequently as the victim of said violence. In the research we are presenting in this PhD Thesis, it is established the importance of the parenting styles in young offenders (the majority due to child to parent violence) and non offenders even though the effect of the relationship with the parents is different in each group. The support of the father and the mother encourages prosocial behaviour and inhibits aggressive behaviour in the offenders group, while inhibiting emotional instability and developing empathy in the non offenders. On the contrary, the parenting style characterised by absent parents who do not attend the needs of the children (negligence) is a risk factor as it contributes to the aggressive behaviour in offenders and reduces empathic concern in non offenders. Extreme permissiveness is also a negative parenting style as it inhibits empathic concern and prosocial behaviour in offenders and contributes to emotional instability in non offenders. In general, parenting styles appear to have a more direct effect on the behaviour of offenders. Aims The general aim of the present work was to analyse the differential profile of young offenders who are carrying out sentences (in their majority due to child to parent violence) and of the non offenders. The profile is focused on emotions and cognitive processes, as well as the two more relevant socialisation contexts, namely the family and the peers. This PhD Thesis is based on data assessment obtained from the six scientific studies that comprise this work, with six specific aims: 1. Given the importance that emotions and emotional instability have in adolescence, the aim of the first study was to analyse the predictor variables for emotional instability in late childhood and the first years of adolescence, taking into consideration personal variables, parenting styles and their relationship with their peers. 2. Continuing the study of emotions in adolescence, and given the impact of anxiety and depression during this time, the aim of the second study was to establish, through a longitudinal study, the variables with higher predictor power of anxiety symptoms and those that discriminate between more or less anxious adolescents. It is of special interest to differentiate between the cognitive processes and emotions that facilitate anxiety symptoms and those that protect the adolescent from said symptoms. 3. Given the co-morbidity between anxiety and depression, the aim of the third study was to analyse the variables greater related to depression, in particular, the interaction between negative affects (emotional instability, anger state and trait, physical and verbal aggression and depression) analysing the different relationships through the time in our sample. 4. From the results obtained in the general population, the following studies focus on comparing how the emotions work and their relation to young offenders and non offenders behaviour. The aim in the forth study was to compare the relations between emotional instability and anxiety; depression and aggressive behaviour modulated by anger in both groups. 5. From the results obtained in the study of emotions it was decided to incorporate cognitive processes. The objective of the fifth study was to analyse the cognitive processes (prosocial moral reasoning, perspective taking) and the emotional processes (empathic concern, emotional instability, anger state-trait) that interact in the prediction of the physical and verbal aggressive behaviour and the prosocial behaviour in young offenders and non-offenders. 6. Even though different studies have analysed the relation between negative emotions and lack of emotional regulation in aggressive behaviour, there is less research that includes cognitive-emotional processes, like emotional awareness and emotional self-efficacy. The aim of this last study was to analyse the mediating role of regulatory emotional self-efficacy and emotional instability in the association between emotion awareness level and aggression in offender populations. Study 1: Emotional Instability in Late Childhood and Early Adolescence: Parenting and Peer Attachment The aim of this study was to analyse the effect of different parenting styles, personal variables (emotional and cognitive), and peer attachment on emotional instability. We paid special attention to gender differences in both parents and children. We studied late childhood and preadolescence (ages 9-12 years). At this age, while parenting style remains important in children’s education and development, children begin to encounter a greater degree of pressure and influence from peers. The participants were 610 students, 316 female and 294 male, with an age range of 9 to 12 years old. They completed the following measures: Children’s Reports of Parental Behaviour: An Inventory (CRPBI) (Samper, Cortés, Mestre, Nácher & Tur, 2006; Schaefer, 1965); The Emotional Instability Scale (IE) (Caprara & Pastorelli, 1993; Del Barrio, Moreno & López, 2001); The Interpersonal Reactivity Index (IRI) (Davis, 1980; Mestre, Frías & Samper, 2004); State and Trait Anger Scale (STAXI) (Del Barrio, Aluja & Spielberger, 2004; Del Barrio, Spielberger & Moscoso, 1998); Attachment to Peers (IPPA) (Armsden & Greenberg, 1987). The results show significant differences between boys and girls. Boys are more emotionally unstable, they experience higher levels of anger and anger externalisation behaviours. In relation to parenting styles they perceive a higher control and hostility in the relationships with their parents, as well as more negligence on their mothers’ part in seeing to their needs. Girls are more empathic (perspective taking, empathic concern and personal distress), they have more self-control mechanisms for anger in situations that produce tension and develop a more sure attachment to peers. The regression analyses show that anger explains 50.9 % of the variance in the prediction of emotional instability in boys, together with a lower degree of empathy, (perspective taking), greater control and hostility in the mother and little peer attachment. The regression analysis of the girls shows that in girls’ externalization of anger, perspective taking and personal distress, and hostility and negligence in the mother, weak self-control, and little affection from the father were predictor variables of emotional instability (35.5% of the variance). The main contribution of this study is to contribute evidence about the prediction of emotional instability from the personal variables, parenting styles of the father and the mother, together with another key factor during development: peer attachment. As for emotions, anger and empathy are related to emotional instability. Moreover, mothers with little involvement in parenting their daughters and who strictly apply rules and punishments, and fathers who offer little affection and emotional support predicted impulsiveness and a lack of self-control. Peer attachment (in the negative) only reaches predictor power in the boys (Bariola, Hughes & Gullone, 2012; Carlo, Mestre, McGinley, Samper, Tur & Sandam, 2012; Groh, Fearon, Bakermans-Kranenburg, van Ijzendoorn, Steele & Roisman, 2014). Study 2: Anxiety in adolescence. Can we prevent it? Taking into consideration the importance of emotions in adolescence, we carried out this study about anxiety. A recent revision of research in anxiety in childhood and adolescence shows that deficits in emotional competence, empathy, emotional self-efficacy, ability to cope in situations that cause tension and generate negative emotions are processes that could be related to anxiety symptoms (Mathews, Koehn, Abtahi & Kerns, 2016; Llorca, Malonda & Samper, 2016). The aim of the present study was to analyse the relation between different representative variables of emotional competency and anxiety in adolescents through a longitudinal study. The participants were 417 adolescents, 192 boys and 225 girls, evaluated over a period of three years. In the first wave the mean age was 14.70 years (SD= 0.68; range= 13-17 years). All of them completed the following instruments in each evaluation: Anxiety and Stress Scales (DASS) (Norton, 2007); Physical and Verbal Aggression Scale (AFV) (Del Barrio, Moreno & López, 2001); Emotional Instability Scale (EI) (Del Barrio, Moreno & López, 2001); State and Trait Anger Scale (STAXI) (Del Barrio, Aluja & Spielberger, 2004); Stress Appraisal Measure for Adolescents (SAMA) (Peacock & Wong, 1990); The Interpersonal Reactivity Index (IRI) (Davis, 1980; Mestre, Frías & Samper, 2004); Adolescent Coping Scale (Frydenberg & Lewis, 1995; Pereña & Seisdedos, 1997). The results again show differences between adolescent boys and girls. The girls score higher in anxiety and empathy (ability to put oneself in the place of others and concern for others), while the boys score higher in emotional instability and physical and verbal aggression. The changes along the evaluated time frame indicate a progressive reduction of aggressive behaviour in both genders and a reduction of emotional instability only in the girls. Both girls and boys reduce their effective coping strategies when facing stressful situations (oriented to the solution of the problem and strategies that lean on seeking support). The hierarchical regression analysis shows that anger state, empathic concern, physical and verbal aggression, unproductive coping mechanisms (ineffective to solve the problem) and the stress perceived as a threat are the variables that stand out in terms of strength to predict anxiety in adolescence. These results give relevant information to prevent anxiety in adolescence. Given that the effective mechanisms to regulate emotions and cope with stressful situations decrease through adolescence and the lack of these mechanisms is related to more anxiety symptoms, it is necessary to train children and adolescents in productive coping strategies, oriented to problem solving, as well as emotional self-control when facing situations that cause tension or require finding a solution to a problem. The reduction in anxiety will contribute to a higher emotional balance and a more adapted behaviour (Barlow, 2000; Grover, Ginsburg & Ialongo, 2007). Study 3: The role of emotions in depression and aggression An increase of emotional disorders like anxiety and depression has been verified in adolescence, also, depressive symptoms frequently present themselves together with anxiety and behavioural problems stemming from anger (Kertz, Belden, Tillman & Luby, 2015; Ingoldsby, Kohl, McMahon & Lengua, 2006). Therefore, in the third study the analysis of the variables related to depression and aggressive behaviour was considered through a longitudinal study. Many people who experience together depression, anxiety, feel irritable or angry, normally cannot interpret nor express their emotional changes appropriately. It is very important to know the interrelation of these emotions very well to be able to help adolescents to manage them more easily. The aim of the third study was double. The first objective was to analyse the interaction between the negative affects (emotional instability, anger state and trait, physical and verbal aggression and depression), analysing their relations through time. The second objective was to provide information to the managing of the negative emotions depending on their relation to other to establish more effective prevention strategies against aggressive behaviour and depression (Andrés, Richaud, Castañeiras, Canet-Juric & Rodriguez-Carvajal, 2016; Li, Xu & Chen, 2015). The participants were 417 adolescents, 192 boys and 225 girls, evaluated for three with a one year gap in between evaluations. In the first wave the mean age was 14.70 years (SD= 0.68; range= 13-17 years). All students completed the following instruments: Physical and Verbal Aggression Scale (AFV) (Del Barrio, Moreno & López, 2001); Emotional Instability Scale (EI) (Del Barrio, Moreno & López, 2001); State and Trait Anger Scale (STAXI) (Del Barrio, Aluja & Spielberger, 2004); CES-D Scale (Radloff, 1991). The results show once again differences by gender in the evaluated variables. Adolescent boys report more physical and verbal aggressive behaviour and more emotional instability throughout the time period evaluated. Depression symptoms increase significantly in boys and girls during the three years. In this case it is the girls who score higher in the three evaluations. The structural models show a difference as to how the emotions work in adolescent boys and girls. Emotional instability is a strong predictor of anger (state-trait) in both genders and anger is a good mediator between emotional instability and physical and verbal aggressive behaviour. Moreover, in adolescent girls emotional instability in the first evaluation directly predicts depressive symptoms in the third wave of the study, meaning, two years later. The longitudinal path analysis show that anger is also a good mediator between emotional instability and depression in girls. These results show that an internalised variable as emotional instability, is relevant in the appearance of depression symptoms in girls and also in the aggressive behaviour of boys and girls to the extent that anger intervenes (Alpaslan, Kocak & Avci, 2016). In regards to anger, there appears a great balance in the role it performs in the case of boys and girls. This involvement means that the control of anger becomes an important goal to control the rest of the negative affects. In the models it becomes apparent that instability predicts depression and anger, but not aggression, the latter only appears if instability is associated to anger. To sum up, the results show important differences in the emotional field between boys and girls throughout adolescence. On the other hand, they provide information about the prediction of depression and aggressive behaviour, information that should be taken into consideration in programmes oriented to establish emotional control in adolescents. Study 4: Depression and aggression in offenders and non-offenders The forth study, as the previous one, focuses on depression symptoms and aggressive behaviour in adolescence, but in this case the study compares young offenders with non offenders. The research on adolescents or youngsters with aggressive or delinquent behaviours shows the importance of various family and social variables (Catrín, Gómez-Fraguela & Luengo, 2015), even though the social factor of aggressive behaviour is one of the factors which explain said conducts, which justifies the study of the internal variables of the individual, especially emotion (Rodriguez, Del Barrio & Carrasco, 2009). Different research has concluded that impulsiveness and lack of emotional control are predictors of anger which in turn determines aggressive behaviour (Mestre, Samper, Tur, Richaud & Mesurado, 2012). Furthermore, recent studies add that emotional instability predicts the appearance of depression symptoms throughout time (Llorca, Malonda & Samper, 2016). Other studies provide information about the relation between delinquency and depression (Ibabe, Arnoso & Elgorriaga, 2014). Therefore, even though there are studies about delinquency and emotions in adolescence, there are less studies on the specific interaction between said emotions. The main contribution of the forth study is to analyse how the negative emotions behave and their relation to maladaptive behaviours in young offenders and if these emotions follow the same pattern as in the general population. The aims of this study were the following: 1) analyse the differences between young offenders and non-offenders in the evaluated variables (emotional instability, anger, aggressive behaviour, anxiety and depression) and also the differences according to gender; and 2) to compare the relation between emotional instability and anxiety, depression and aggressive behaviour mediated or modulated by anger in both groups. The participants were 440 adolescents, 220 of those were young offenders who were selected from four different correctional centres of the Valencia Region, who where carrying out different court sentences. Most of them were carrying out a sentence due to child to parent violence. The rest of the sample was selected randomly from ten public and private schools in the Valencia metropolitan area. The selection was made equating both subgroups in age, gender and controlling the representation of the social status, verifying that there aren’t significant differences between the two groups related to these socio demographic variables. The subgroup of young offenders includes a total of 148 boys and 72 girls; amongst the participants selected from the general population we find a total of 145 boys and 75 girls. The ages of the subjects are between 15-18 years. Both groups of adolescents filled in the following instruments: Physical and Verbal Aggression Scale (AFV) (Del Barrio, Moreno & López, 2001); Emotional Instability Scale (EI) (Del Barrio, Moreno & López, 2001); State and Trait Anger Scale (STAXI) (Del Barrio, Aluja & Spielberger, 2004); CES-D Scale (Radloff, 1991; Eaton, 2004); Anxiety and Stress Scales (DASS, Norton 2007). The results show significant differences between the two groups in all variables evaluated, being the young offenders the ones to reach higher scores in anxiety, depression, anger, emotional instability and physical and verbal aggressive behaviour. Likewise, gender differences are established: the girls present more depression symptoms in both groups, only in the non offender sample a higher level of aggressive behaviour can be observed in the boys, while in the young offender sample there is no difference, both boys and girls reach higher levels of aggressive behaviour. This result confirms higher levels of physical and verbal aggressive behaviour in young offenders independently of the gender of the adolescent who has committed the crime. The structural models show that in the young offender’s group emotional instability is a direct predictor of anger trait and aggressive behaviour. Anger also mediates anxiety and depression, in turn anxiety is strongly related to depression. In the non offender adolescents the structure is different, emotional instability relates directly to anger, aggressive behaviour and depression, while anger does not relate to anxiety or depression. Anxiety also directly relates to depression although not as strongly as in the offenders group. A relation between physical and verbal aggressive behaviour and depression symptoms is not established neither in the young offenders nor the non offenders. This result does not coincide with other studies that conclude a co-occurrence of aggressive behaviour with depression symptoms in adolescence (Van der Giessen, et al., 2013). To sum up, we conclude that in the offender subjects all emotions are closely related to each other, as well as the important role of anger in anxiety and depression. This data highlights the need to act upon the offenders group with the aim to improve their emotional situation. The target emotions would be those which we find in the initial explanatory core and they are emotional instability and anger trait. The present study provides relevant results about the emotional profile of young offenders and non offenders, about how the negative emotions behave and their relation to maladapted behaviour. And on the other hand, if said emotions follow the same pattern as in general population. The results widen the knowledge of the different role of of anger in both groups. These results have consequences over prevention. However it is important to take into account the temperamental dimension of emotional instability and consider the importance of emotional control of impulsive children, it is basic to intervene in the management of regulation of anger as emotion when it appears in response to adverse or frustrating situations. Study 5: Prosocial reasoning and emotions in offender and non-delinquent adolescents In keeping with the aim to establish a differential profile between offenders and non offenders, the fifth study incorporates the analysis of cognitive processes together with emotional processes. Different studies establish that a negative emotionality together with an inability to regulate emotions predict antisocial and delinquent behaviour (Caprara, Gerbino, Paciello, Di Giunta & Pastorelli, 2010; McMahon et al., 2013). Conversely, empathy is considered an inhibitor factor of aggressive and delinquent behaviour (Carlo et al., 2010; Van der Graaff, Branje, De wied & Meeus, 2012), even though the results on the relation of empathy, aggressive behaviour and delinquency are inconclusive. Although there are studies that confirm a negative relation between empathy and aggressive behaviour (Jolliffe & Farrington, 2004) and they consider empathy as a protection factor from aggressive behaviour (Carlo, et al., 2010; Wang, Lei, Yang, Gao & Zhao, 2016), other studies do not find significant differences between offenders and non offenders in the cognitive and affective factors of empathy (Schalkwijk, Jan Stams, Stegge, Dekker & Peen, 2016). On the other hand, the research on moral conduct has highlighted the need to include moral cognitions together with emotions when explaining said behaviour. This justifies the inclusion of moral prosocial reasoning in the study of adapted and maladapted behaviour, in particular to explain the prosocial behaviour (behaviour oriented to benefit others) versus aggressive behaviour (behaviour oriented to harm others). Even though there is a vast research on empathy, moral prosocial reasoning and prosocial behaviour, the relation between moral prosocial reasoning and aggressive behaviour has been less studied and even less how these cognitive processes interact with empathy and with negative emotions such as anger and lack of self-control in offender population. The main objective of this study was to analyse the cognitive processes (moral prosocial reasoning and perspective taking) and the emotional processes (empathic concern, emotional instability, anger state-trait) which interact in the prediction of aggressive behaviour and prosocial behaviour of young offenders and non offenders. The purpose of this analysis is to establish a differential profile, according to the predictor variables in both groups, which will help prevent delinquent behaviour. The result will conclude if aggressive and prosocial behaviour perform differently in both groups of adolescents and if to explain these behaviours the reasoning processes as well as empathy and emotional regulation should be considered. Other specific aims were focused on analysing the differences between cognitive and emotional processes depending on gender and between the young offenders and non offenders. The participants were 440 adolescents, 220 of those were young offenders who were selected from four different correctional centres of the Valencia Region, who where carrying out different court sentences. Most of them were carrying out a sentence due to child to parent violence. The rest of the sample was selected randomly from ten public and private schools in the Valencia metropolitan area. The selection was made equating both subgroups in age, gender and controlling the representation of the social status, verifying that there aren’t significant differences between the two groups related to these socio demographic variables. The subgroup of young offenders includes a total of 148 boys and 72 girls; amongst the participants selected from the general population we find a total of 145 boys and 75 girls. The ages of the subjects are between 15-18 years. Both groups of adolescents filled in the following instruments: Prosocial Reasoning Objective Measure (PROM) (Carlo, Eisenberg & Knight, 1992; Mestre, Frías, Samper & Tur, 2002); The Interpersonal Reactivity Index (IRI) (Davis, 1980; Mestre, Frías & Samper, 2004); Prosocial Behavior Scale (PB) (Caprara & Pastorelli, 1993; Del Barrio, Moreno & López, 2001); Physical and Verbal Aggression Scale (AFV) (Del Barrio, Moreno & López, 2001); Emotional Instability Scale (EI) (Del Barrio, Moreno & López, 2001); State and Trait Anger Scale (STAXI-N) (Del Barrio, Aluja & Spielberger, 2004). The results show a differential weight of the reasoning processes, empathy and emotional regulation in the explanation of said conducts, as well as differences by gender and between young offenders and non offenders in the evaluated processes. Young offenders use hedonistic reasoning (arguments focused on the personal benefit of the action) more than the non offenders, as well as the reasoning oriented to seek approval from others, furthermore, they are more emotionally unstable and show more anger and aggressive behaviour. On the contrary, the non offender adolescents are more empathic and more prosocial. The differences between boys and girls can be seen in empathy. The girls of both groups (offenders and non offenders) have a higher capacity to put themselves in the place of another than the boys of the same age. In the same way, only the non offender girls report more empathic concern and more prosocial behaviour than the boys of their group. On the other hand, it is the non offender boys who reach higher scores in aggressive behaviour, compared with the girls in their group; whereas these differences disappear in the offenders group. There is no differences according to gender in any of the moral prosocial reasoning categories evaluated. The regression analysis for the prediction of physical and verbal aggressive behaviour of the young offenders and non offenders explains a high percentage of variance (58% in offenders and 46% of variance in the general population). The scarce capacity to put oneself in the place of another, emotional instability and anger (state) are the variables which better predict the aggressive behaviour in offender population. In the non offenders group, the hedonistic reasoning and the reasoning that seeks the approval from others together with low empathic concern and emotional instability intervene in the prediction. With regards to the prediction of prosocial behaviour, there are also differences between both groups. Even though empathy (perspective taking and empathic concern) appears as a strong predictor in the offender and non offender samples, it is anger in the negative sense the variable that also appears in the prediction in the offenders group. On the other hand, in the non offender group, the stereotyped reasoning comes into the prediction of prosocial behaviour. The results show that prosocial reasoning has a lesser weight in the prediction of aggressive behaviour and of prosocial behaviour, especially in the offender population, even more if it is compared to empathy, which is the strongest predictor of these behaviours in both groups. On the contrary, emotional instability is one of the variables with stronger predictor power of aggressive behaviour. These results are relevant for the prevention and the re-education programmes oriented to young offenders. The strategies of emotional regulation, the development of empathy, in both its cognitive component (understand the state of another and to put oneself in their place), and its affective component (feelings oriented toward another) must be present in theses programmes. Also, given that moral prosocial reasoning does appear as predictor of prosocial behaviour and inhibitor of aggressive behaviour in the adolescent general population, and it is absent in the offender sample, programmes for the training on moral prosocial reasoning should be included, oriented to internalise principles and arguments oriented to empathy, the common good, respect for the norms and less oriented to hedonism or to seek approval from others. Study 6: Emotional awareness and aggression in adolescence: the role of emotional instability and regulatory emotional self-efficacy beliefs. Finally, to complete the study of emotions, a study oriented to emotional competence and self-efficacy was designed. Even though different studies have analysed the relations between negative emotions and lack of emotional regulation in aggressive behaviour, as the already mentioned articles conclude, there is less research that shows the cognitive-emotional processes such as emotional awareness and emotional self-efficacy as vulnerability or protection variables of aggressive behaviour. With the aim of reaching further in the objective of this PhD thesis, the present study was also carried out with a sample of young offenders and non offenders with the objective to analyse if these processes operated differently in both groups. Emotional awareness defined as an attentional process which implies interpretative and evaluative functions, this process allows the control of our own emotions, distinguishing them and identifying their causes and physiological correlations (Rieffe, Oosterveld, Miers, Terwogt & Ly, 2008). Emotional self-efficacy refers to the believe in the ability a person has to improve negative emotional states which are activated when facing stressing situations, the ability to prevent negative emotions such as anger, irritability, or depression, as well as the perceived ability to express positive emotions like happiness or enthusiasm(Caprara et al., 2008; Caprara, Vecchione, Barbaranelli & Alessandri, 2013). From previous research the main aim considered is to analyse the mediator role of emotional self-efficacy and emotional instability between emotional awareness and aggressive behaviour and anger. It is a question of testing this model in offender and non offender populations. Data was obtained from 245 male and female inmates (average age = 16.25) across four Youth Detention Centres in the Valencian Community, Spain, centres in which the minors serve their sentences for crimes committed (more than 60 % were carrying out a sentence due to child to parent violence), and 416 male and female community participants (mean age = 16.73). The community participants were obtained from public and private schools in the Valencian Community, participating schools were randomly selected from the list of all schools in Valencia with students enrolled in compulsory secondary education. Participants were asked to complete a series of measures designed to assess their emotional awareness, regulatory emotional self-efficacy, emotional instability and different scale to measure aggression. All of them filled in the following instruments: Emotion Awareness Questionnaire (EAQ30; Rieffe et al., 2008, Spanish version Samper-García, Mesurado, Llorca, & Richaud, 2016; Regulatory emotional self-efficacy (Caprara, et al., 2008); Emotional Instability Scale (IE, Caprara & Pastorelli, 1993; Del Barrio, Moreno & López, 2001); Physical and Verbal Aggression Scale (AFV, Caprara & Pastorelli, 1993; Del Barrio, Moreno & López, 2001); State and Trait Anger Scale (STAXI-N) (Del Barrio, Aluja & Spielberger, 2004). The results, once more, show the differences between both groups in the evaluated variables. The young offenders report lower levels of emotional awareness and emotional self-efficacy in the self-efficacy dimensions in expressing positive emotions and self-efficacy in controlling anger and irritability. On the other hand, these adolescents reach higher levels of aggressive behaviour (physical and verbal) and anger (state-trait). The mediation analyses carried out indicate that the association between emotional awareness and aggressive tendencies can be due, at least in part, to emotional self-efficacy. In the offender sample the results indicate that those with lower levels of emotional awareness report more aggressive behaviour and more anger, however the mediation hypothesis is not confirmed. Therefore, there is no indirect effect of the level of emotional awareness on aggressive behaviour and anger through emotional self-efficacy. The results follow this line in the general population. As to the mediator role of emotional instability between emotional awareness and aggressive behaviour, the results indicate that there are no indirect effects of the level of emotional awareness on different aspects of aggressive behaviour and anger (physical, verbal, anger state, anger trait) through emotional instability in the offender population. On the other hand, in the non offender adolescents can be observed a direct effect of the level of emotional awareness on the different aspects of aggressive behaviour evaluated, through emotional instability, even though the effect of the mediation is partial. These results widen the knowledge of the cognitive-emotional variables that can be related to externalised behavioural problems. By way of conclusions, it is established that non offender adolescents show more emotional awareness, meaning, they are more able to identify and differentiate emotions, as well as their origin. Moreover, they show more self-efficacy to manage positive and negative emotions. On the other hand, the young offenders show that they are more emotional unstable. Emotional self-efficacy appears to have a protection effect towards physical and verbal aggressive behaviour and anger states in the general population, no in the offender population. Furthermore, emotional awareness prevents emotional instability only in the general population, but not in the offender population. On the other hand, emotional awareness maintains its protection of physical and verbal aggressive behaviour and anger state in both samples, but only of anger trait in the general population. These results show that the analysis of the emotional world is necessary to prevent different manifestations of aggressive behaviour, even when the levels of aggressive behaviour are high and they result in maladapted behaviour as is the case of the offenders. When the mediator power of emotional instability is analysed, it is observed that emotional awareness has a direct and total effect over each of the aggressive behaviour variables analysed in both samples (with the exception of anger trait, the effect of which is only given in the general population). We could think that the aggressive behaviour in the offender population can be better explained from more stable character traits, cognitive-emotional abilities such as emotional awareness. We can assume that emotional instability cancels out the ability to identify emotions not being able to stop the escalation of disruptive aggressive behaviour. Discussion These six studies provide strong evidence of the importance of emotions in adolescence and their implication in the adapted and maladapted behaviour, particularly in the prosocial and aggressive behaviour. The positive and negative emotions of young offenders and non offenders have been evaluated. Based on the research on this subject in recent years. The results have highlighted the importance of empathy to inhibit aggressive behaviour and delinquency, on the othe hand anger appears as the target emotion in maladapted social behaviour. In young offenders all emotions are closely related to each other, but the target emotions would be those present in the initial explanatory core and they are emotional instability and anger trait. This data makes the need to act on the young offender group obvious, with the aim to improve their emotional situation and their ability to self-control. Even though the moral reasoning processes have a secondary place in young offenders, they have to be taken into account because they predict the aggressive and prosocial behaviour in non offenders. Therefore, developing different arguments aimed at personal benefit or arguments aimed at the benefit of others, or at the respect for the established rules, together with emotional regulation take the adolescent towards personal independence and they can contribute to the prevention of maladapted behaviour. To sum up, this research has practical implications, especially to develop programmes aimed at increasing empathy and prosocial behaviour and to prevent aggressive behaviour. Also for the re-education of young offenders through the development of empathy, emotional regulation, anger in particular, emotional awareness and self-efficacy, not forgetting moral prosocial reasoning aimed at personal independence and the respect for the rules, reasoning which is almost absent in the offender population. A greater understanding of the emotions involved in aggressive behaviour can help in the design of more effective prevention and intervention programmes that encourage abilities and strategies to inhibit said behaviour and boost prosocial behaviour through empathy and emotional regulation. Even though the regulation of emotions is at the heart of prevention of aggressive behaviour and delinquency, we cannot forget the role of the parents. It is key that children grow in a family environment characterised by affection, communication and rules. Permissiveness and negligence are the most negative parenting styles in the socialisation progress. Therefore, it is crucial, the simultaneous re-education of parents and children when facing delinquent behaviour, in particular with regards to child to parent violence.