Background: Clark and Watson proposed a tripartite model in which symptoms of anxiety and depression in adults are viewed along three broad dimensions. The first, general affective distress or negative affect (NA), is associated with both depression and anxiety; the second, physiological hyperarousal (PH), is specific to anxiety; and the third, a lack of positive affect (PA), is specifically associated with depression. Nonspecific shared symptoms of NA are thought to explain the strong association between measures of anxiety and depression. Negative Affect reflects the extent to which a person is experiencing negative mood states such as fear, sadness, anger, and guilt, whereas Positive Affect reflects the extent to which one reports positive feelings such as joy, enthusiasm, energy, and alertness. Physiological hyperarousal includes somatic tension and hyperarousal (e.g. shortness of breath, dizziness and light headedness, dry mouth) Although formulated with adults, the tripartite model also has important implications for the nature, classification, assessment, and treatment of anxiety and depression in children. A limitation of Clark and Watson's model is that it focuses primarily on affective and physiological symptoms, to the relative neglect of cognitive aspects of anxiety and depression. Fuller incorporation of cognitive phenomena into the tripartite model has important potential benefits. Because several theories of anxiety and depression emphasize the importance of cognitions, identifying cognitive features common and specific to each would enhance our understanding of and ability to differentiate between them. In addition, given the prominence of cognitive approaches to the treatment of anxiety and depression, identifying cognitions associated with each would aid in development and refinement of effective interventions. Thus, our study examined aspects of cognition common and specific to anxiety and depression. According to the Cognitive Model cognitions determine individual’s emotions and behaviour. Tripartite and Cognitive Model have limitations that emanate from their focus on only one aspect of disorders, whether emotional as a basis of the Tripartite Model or cognitions as a foundation of the Cognitive Model. Therefore, the main goal of this paper is to improve differentiation between anxiety and depression through combining both models. Aim: The aim of this study is to verify the compatibility between the Tripartite and Cognitive Model through their integration, while observing the specificities related to gender and severity of disorder. Method: The aim of this study is to verify the compatibility between the Tripartite and Cognitive Model through their integration, while observing the specificities related to gender and severity of disorder. Method: Participants In non-clinical sample participants were sixth-grade elementary school students at the age of 12 years from three schools selected for their socioeconomic and geographic diversity in Zagreb. Only children who returned affirmatively signed parental consent forms were allowed to participate in the study. Parent consent and child assent were obtained for 350 children. 33 students failed to complete four or more items and their data were not included in the analyses. The final sample included 317 children (154 boys, 163 girls). The control group criteria was that the child achieved result of M or lower on the depression or anxiety questionnaire. A sample of 150 children were randomly selected according to this criteria, 70 boys and 80 girls. Testing in clinical groups was conducted at Psychiatric hospital for children and youth during a preriod of two years, on a conventional sample of participants divided into two groups: one consisting of clinically depresed subjects and other consisting of clinically anxious subjects. All children were early adolescents. In total, 110 children were selected from the patients of the Psychiatric hospital for children and adolescents, and additional 46 children were recruited from the sample of 350 children, which have participated in the study conducted during the academic year 2008/09. The inclusion criteria for the children in the clinical group was that the child achieved result of M+1SD or greater on the depression or anxiety questionnaire. A total of 156 children were included in the clinical group. Anxiety clinical group consisted of a total of 80 children, 37 boys and 43 girls, while the depression clinical group consisted of 76 children, 40 boys and 36 girls. Procedure Research was conducted in accordance with the Ethical code for research involving children.Investigator visited classrooms to invite students to participate in "a study of children's thoughts and feelings about everyday situations" and to distribute consent forms for students to take home to their parents. Only children who returned affirmatively signed parental consent forms were allowed to participate in the study. Group testing lasted 45 minutes (one school period). Children in clinical group were invited to take part in the study as a part of pychological examinationm, psyhciatrical examination or at the begining of a psychiatrical treatment. Parents who agreed to their child taking part in the study signed a consent form. In the Psychiatric hospital for children and youth the testing was performed individually. As in examination in control group, the examiner read the instructions and children noted their answers and asked for clarification if needed. On each questionaire childen noted their age and sex, and after conclusion the examiner noted into which clinical group the child belongs (anxiety or depression). The diagnosis was made by a psychiatrist based on diagnostic criteria cited in the MKB-10. In the selection of participants care was taken that comorbidity of anxiety and depression disorder was not present. Measures State-Trait Anxiety Inventory for Children (STAIC) The trait scale of the STAIC was used as a measure of general anxiety (Spielberger, 1973). This scale has been reliably used with children ranging in age from 6 to 16 years. Each of 20 items is scored from 1 (Hardly ever) to 3 (Often true), with higher scores indicating greater anxiety. Coefficient alphas of .81 and .78 along with test–retest reliabilities of .71 and .68 have been reported for girls and boys, respectively. Children’s Depression Inventory (CDI) The CDI was used to assess symptoms of depressed mood (Kovacs, 1980/81, 1992). This 27- item self-report measure has acceptable reliability (α = .86) and 1-month test–retest reliability (r = .72), and is suitable for use with children and adolescents aged 7–17 years. A variety of symptoms of depression are assessed (e.g., sleep disturbance, suicidal thoughts, and general dysphoria). Each item consists of three brief statements that describe a range of responses scored as 0, 1, or 2, with higher scores indicating increasing severity (0 denotes an absence of symptom, 1 denotes a mild symptom, and 2 denotes a definite symptom). Respondents rate the degree to which each group of statements describes their mood during the previous 2 weeks. Positive and Negative Affect Scale for Children (PANAS-C) The PANAS-C is a 27-item scale designed to assess NA and PA in children (Laurent et al.,1999). It includes 12 items on the PA scale and 15 items on the NA scale. Children are asked to rate the degree to which different words (e.g., “interested,” “sad”) describe how they have felt during the past few weeks, on a scale of 1 (Very slightly or not at all) to 5 (Extremely). Laurent et al. (1999) reported acceptable alpha coefficients (.94 and .92 for NA, and .90 and .89 for PA) for the scale development and replication samples, respectively. Good convergent and discriminant validity were also reported, with the NA scale correlating positively with self-reports of depression (r = .60) and anxiety (r = .68), and the PA scale correlating negatively with depression (r = −.55) and to a lesser extent with anxiety (r = −.30). Physiological Hyperarousal Scale for Children (PH-C) The PH-C was developed to complement the PANAS-C, thereby assessing all three components of the tripartite model (Laurent, Catanzaro, & Joiner, 2000). It is an 18-item measure of PH, defined as bodily manifestations of autonomic arousal. Respondents are asked to rate the degree to which different words (e.g., “dry mouth,” “sweaty palms”) describe how they have felt during the past 2 weeks, on a scale of 1 (Very slightly or not at all) to 5 (Extremely). The measure was developed with a nonclinical sample of students in grades 6 through 12. A coefficient alpha of .86 was reported by Laurent et al. (2000). Children's Thought Questionnaire (CTQ) CTQ was used to assess children's self-reported anxious, depressive, and positive thoughts (Marien i Bell, 2004). Ten vignettes, each followed by a series of corresponding thought and mood rating items, were presented in the same order to all children. Vignettes were written to reflect common experiences that are challenging to middle school age children in general, including those with a tendency to worry or experience of anxiety or depression. Vignettes addressed school (e.g., tests, grades), social evaluation (e.g., being criticized or rejected by peers, being called on in class), and health issues. Each vignette is followed by a series of six thought items that were designed to represent anxious (two items), depressive (two items), and positive (two items) cognitive content. Authors wrote items to be consistent with theoretical and empirical literature on cognitive content characteristic of anxiety versus depression. Positive thought items reflect the expectation of a positive outcome, interest or pleasure, or positive self-reference (e.g., "I'm good at this subject"). Using a 5-point Likert scale ranging from 1 (not at all like I would think) to 5 (exactly like I would think), children rate the similarity of each cognition to thoughts they would have if they were in the hypothetical situation. A coefficient alpha in our study was .835. Results: Girls reported more anxiety, more negative cognitions and negative affectivity less positive cognitions than boys. No gender differences were found in depression, positive and and symptoms of physiological hyperarousal. Negative affectivity is a significant predictor of anxiety and depression in non-clinical sample. In clinical sample negative affectivity is specific to depression. Lack of positive affect was a significant predictor of depression in both non-clinical and clnical samples. Contrary to the tripartite model, in clinical sample our results showed that negative affect is specific to depression. Physiological hyperarousal was a significant predictor of depression in both non-clinical and clnical samples. Negative and positive cognitions explain additional share of variance in anxiety and depression. Correlation between positive cognitions and depression is partly mediated with positive affect, correlation between negative cognitions and depression is partly mediated with negative affect, correlation between negative cognitions and anxiety is partly mediated with negative affect and physiological hyperarousal. Unexpected finding in this study was that physiological hyperarousal is also predicted by a positive cognitions. Based on these results, we can conclude that the theoretical model explaining anxiety and depression may be expanded to integrate the tripartite and cognitive models. The integrated model described fits equally well to all groups, regardless of sex, health or clinical group. Conclusion: The scientific contribution of this research is improved differentiation between anxiety and depression in early adolescence through a clearer insight into the relationship between emotions, physiological symptoms and cognitions related to the two disorders. Aldough the results do not show differences in how the model fits in regards to sex or presence of psychopathology, age- and gender-specific models to better account for the shared and unique aspects of depression and anxiety in children need to be further explored.