Carisbamate (CRS) and lamotrigine (LTG) are anticonvulsants which act mainly on neuronal voltage-gated sodium channels, that have been shown to have antidepressant-like effects in animal models of depression. In vivo electrophysiological recordings were carried out following 2 and 14 days of CRS or LTG administration. Overall firing activity in the dorsal raphe, locus coeruleus and ventral tegmental area were decreased with CRS. Similarly, a decrease in the dorsal raphe was also observed with LTG. Despite these presynaptic decreases in firing activity, both anticonvulsants exhibited significant enhancement of serotonergic transmission in the hippocampus as demonstrated by increased tonic activation of postsynaptic 5-HT1A receptors. This may be attributed to the observed desensitization of the terminal 5-HT1B autoreceptors. This study suggests that the enhanced serotonergic effect may be associated with an antiglutamatergic effect, and may contribute to the antidepressant-like effect of CRS in the forced swim test and the antidepressant properties of LTG.

Background: Understanding of child development is predominantly based on maternal influences on children’s emotional, behavioural and cognitive outcomes. Although there has been an increase in research focus on fathers in recent years suggesting that fathers are important in the development of their child, there is still a shortage of research on fathers in the literature. Research has shown fathers negatively impact on their children’s emotional, behavioural and cognitive development, but there is a lack of understanding regarding the specific mechanisms through which paternal depression influences their children. The aim of the current PhD is to address this gap in the literature and this is done by: a) investigating the prevalence of depressive symptoms among fathers of children (aged 9 months – 7 years) and their associated risk factors; b) investigating the association between paternal depressive symptoms and different aspects of parenting such as warmth, conflict and involvement; c) testing whether fathers’ parenting mediated any association between paternal depressive symptoms and children’s emotional, behavioural and cognitive outcomes; and d) investigating the association between paternal depression/depressive symptoms and insecure father-child attachment. Methods: There are two methods employed for this PhD. One is secondary data analysis of the large and representative Millennium Cohort Study (MCS) (investigating a, b and c) and the other is primary data analysis of the Fathers-in-Focus (FIF) study using interview and observational methods to investigate (d). Results: Paternal depressive symptoms peak during the first year of children’s lives and then gradually decline between the first year and 7 years old (a). These depressive symptoms across the first 7 years of fatherhood were consistently associated with maternal depressive symptoms, relationship conflict and unemployment (a). Moreover, depressive symptoms in the first year were associated with father-child conflict, but not father-child warmth or involvement in parenting activities (b). Father-child conflict mediated the association between paternal depressive symptoms and children’s emotional and behavioural outcomes (c). Finally, father’s depressive symptoms were not associated with father-child attachment or children’s cognitive development (c and d). Conclusion: The key finding of this PhD is that father-child conflict is an important factor that may be associated with the risk transmission of paternal depressive symptoms and children’s emotional and behavioural outcomes. Therefore, it may be beneficial for service providers and clinicians to target interventions with depressed fathers’ and at-risk families.

Zusammenfassung: Hintergrund: Mit einer Lebenszeitprävalenz von 1 % und einem Ersterkrankungsalter in der frühen Adoleszenzperiode verursachen Schizophrenien (SCZ) und bipolare Störungen (BPD) großes individuelles Leid. Die genetische Komponente beider Erkrankungen liegt mit einer Heritabilität von bis zu 80 % im Vergleich zum Anteil von Umweltfaktoren sehr hoch. Aufgrund seiner Lage auf dem Locus 12q22-24, einem Hot spot für SCZ und BPD, stellt DYNLL1 ein interessantes positionales Kandidatengen dar. Das Protein, eine 8 kD schwere leichte Dyneinkette ist ein multifunktionales Protein. Durch seine Funktion als Inhibitor von NOS-I, seiner Beteiligung am postsynaptischen NMDA-Proteinkomplex, einer möglichen Interaktion mit NUDEL/DISC1, seiner Ähnlichkeit zu KIF2 und nicht zuletzt wegen der Interaktion mit KIBRA stellt DYNLL1 auch aufgrund seiner Funktion ein relevantes funktionelles Kandidatengen für beide Erkrankungen dar. Methoden: In einer Fall-Kontrollstudie wurden daher sechs Single nucleotid polymorphismen (SNPs) und deren entsprechenden Haplotypen bei 284 Kontrollen, 246 Patienten, die an einer SCZ und 90 Patienten, die an einer BPD litten analysiert, um eine Assoziation dieses Gens mit den entsprechenden Phänotypen zu untersuchen. Ergebnisse: Es zeigte sich eine Assoziation des Markers rs787828 mit SCZ, darüber hinaus eine signifikante Assoziation eines Haplotyps (TTATAG), letztere allerdings nur mit einer Frequenz von 1%. Bei der bipolaren Störung waren dagegen sowohl zwei Polymorphismen (rs1167705 und rs580016), als auch ein Haplotyp (TTGTAG) signifikant mit der Erkrankung assoziiert. Aufgrund der kleineren Stichprobengröße ist es jedoch wichtig, diesen Befund nochmals zu replizieren, um falsch positive Befunde auszuschließen. Zusammenfassung: Sowohl SNPs als auch Haplotypen im DYNLL1 Gen zeigten Assoziationen mit Schizophrenie und der bipolaren Störung, was die These unterstützt, dass DYNLL1 ein relevantes Kandidatengen für beide Erkrankungen ist. Um die Bedeutung von DYNLL1 in der Pathophysiologie der SCZ und der bipolaren Störung weiter aufzuklären, müssen die assoziierten Varianten bezüglich möglicher Auswirkungen auf Genexpression, Proteinfunktion und physiologische Parametern hin weiter untersucht werden. / Summary: Background: Schizophrenia and bipolar disorder, two of the most devastating mental illnesses, have a substantial genetic background with an heritability of up 81%. The gene encoding DYNLL1 is located at 12q 22-24, which represents a major linkage hot spot for these disorders. DYNLL1, an 8 kD dynein light chain, is a multifunctional protein which inhibits all NOS isoenzymes, interacts with the NMDA protein complex and possibly with the NUDEL / DISC 1 proteins. Due to its effect on nitric oxide (NO) synthesis as well as its chromosomal localization, it is considered a promising candidate molecule in the pathogenesis of schizophrenia and bipolar disorder. Methods: Six single nucleotid polymorhisms (SNPs) in the DYNLL1 gene have been genotyped by primer extension and MALDI-TOF analysis in 264 patients suffering from chronic schizophrenia, 90 patients with bipolar disorder and 286 healthy controls. Subsequently, associations for single makers, as well as the corresponding haplotypes were tested for. Results: Single marker association analysis showed that one SNP (rs787828) and one haplotype (TTATAG) were linked to schizophrenia. However, this haplotype is rare with a frequency of only 1%. Two SNPs (rs12857 and rs580016) and one haplotype were associated with bipolar disorder, yet it has to be considered that this sub-sample is very small. Conclusions: Single markers as well as haplotypes in the DYNLL1 Gene were associated with schizophrenia and bipolar disorders, further underscoring the notion that DYNLL1 is a relevant candidate in the pathogenesis of these disorders. Given the preliminary nature of the findings, further studies however are clearly warranted to clarify its role in mental disease.

Introduction: Health professionals performing their social health service (SHS) in rural communities could be at risk of developing depression. Moreover, those who migrate from farther places to perform their SHS could have an increased risk. The objective of this study was to evaluate the association between place of origin and the presence of depressive symptoms, in health professionals performing rural social health service (SHS) in Ancash, Peru. Methods: This was a cross-sectional study. During April 2015, a survey was applied to health professionals performing SHS in the Peruvian Ministry of Health (MINSA) facilities in Ancash. The main outcome was the presence of depressive symptoms, defined as a score =2 points in the Patient Health Questionnaire-2. The main exposure was the place of origin, defined as the place where the subjects completed their undergraduate professional studies (Ancash, Lima city or others). Poisson regressions with robust variance were performed to calculate crude and adjusted prevalence ratios (PR and aPR) and their 95% confidence interval (95%CI). Results: From 573 health professionals performing their SHS in MINSA in Ancash, 347 were included in the study. The mean age was 27.2±4.5 years, 78.7% were women, and 14.7% scored positive for depressive symptoms. Those who had completed their undergraduate professional studies in Lima city had a higher prevalence of presence of depressive symptoms compared to those who did in Ancash (aPR=2.59, 95%CI=1.23-5.45). Conclusions: Those who completed their undergraduate professional studies in Lima had a higher prevalence of depressive symptoms than those who did in Ancash. Possible explanations include the difficulty in visiting family and friends, acculturation, and lack of Quechua language proficiency. / Revisión por pares

This PhD investigated the efficacy of individual therapies for depression and went on to test metacognitive therapy (MCT) for major depressive disorder (MDD) in individual therapy and in transdiagnostic groups consisting of a range of disordersStudy 1 included a systematic review of meta-analyses comparing the effects of individual therapy for MDD across studies. The findings show small to moderate effect sizes between g=0.25 to d= 0.69 and recovery rates 34% to 47.9% for ITT analyses. However, studies are biased and lack objective definitions of recovery, remission and clinically meaningful change which makes comparisons across studies challenging. Study 2 aimed to test MCT in a single case study with four depressed Danes in an outpatient setting. Three out of four patients reached recovery levels (BDI-II smaller or equal to 8) in only five to eleven sessions and all four patients were recovered at 6-months follow-up. Study 3 involved a large randomised clinical trial (n= 153) in which the effect of MCT was compared to cognitive behaviour therapy (CBT) for MDD. Patients were allocated to up to 24 sessions of treatment and were assessed at pre, post and 6 months follow-up on primary and secondary measures. The mean number of sessions were significantly lower for MCT (5.5; SD = 2.4 versus 6.7; SD = 4.7) and MCT showed a higher completion rate (73.6% versus 65.4%). Both treatments were associated with significant improvements in depression measured with the HDRS and BDI-II. MCT was superior in its effects on the BDI-II and on secondary measures, showing a clear advantage of MCT. . Large ES were detected in both MCT and CBT. Using Jacobson and Truax (1991) criteria revealed that 76% reached recovery levels at post-treatment in MCT whereas 54% reached recovery in CBT. These findings were maintained for both conditions at 6-months follow-up. Study 4 evaluated the effect of MCT in a 6-week treatment protocol for mixed groups of diagnosis in an open trial (n= 131). Significant improvements were observed in outcomes and 85% of patients were reliably improved at post-treatment as measured on the HADS. These findings were maintained at follow- up and the treatment appeared effective in both anxious and depressed cases. In conclusion existing treatments for depression are effective but there is much room for increasing efficacy. MCT appeared more effective than a current treatment of choice; CBT in depression. MCT was also associated with significant improvement in anxiety and depression in patients in a transdiagnostic group setting. The results support the future study and implementation of MCT as an effective treatment option.

Social defeat is an ethologically relevant stressor that utilizes the natural establishment of social rank in male rodents and has been shown to be relevant to major depressive disorder (MDD) and post-traumatic stress disorder (PTSD). In the present study, we wished to establish a social defeat stress model in combination with the chronic unpredictable stress model, which is considered a mild stressor to the rodent. In this way, we create a “double hit” model that may more accurately mimic severe stress that is common in both MDD and PTSD. In the present study, residents established dominance over the intruder for 10 consecutive days. In addition, social defeat stress was followed by another stressor given at random times during each day, i.e. chronic unpredictable stress. These unpredictable stressors included 30 min restraint, 1 h shaking/crowding, a cold water swim, a warm water swim or a tipped cage for 24 h. In one cohort of animals, brain tissue was taken 24 h after the last stressor for DNA. In a second cohort, animals were tested on a sucrose preference test in which two bottles containing 0.8% sucrose was placed on their cages for 3 consecutive days (days 8-10 of social defeat stress), and the total amount of sucrose was calculated relative to total volume consumed. Brain tissue analyses revealed significantly elevated DNA oxidation in white matter comparing stressed animals to non-stressed controls, consistent with what has been found in post-mortem white matter from MDD subjects. Further, animals given the social defeat + chronic unpredictable stress demonstrated a deficit in sucrose preference, a natural reward, revealing that these animals were anhedonic as compared to controls. Stressed animals also demonstrated fear of the intruder in a social interaction test performed one day after the social defeat/chronic unpredictable stress was complete. Therefore, it appears that social defeat plus chronic unpredictable stress produces a phenotype relevant to clinical data in humans.

Research has shown that sleep disturbance (e.g., sleep latency, wakefulness after sleep onset [WASO]) negatively affects physical, emotional, and mental health. For many adults, the postpartum period is one that is particularly highlighted by sleep disturbance and fatigue. Postpartum mothers are also more vulnerable to psychiatric disorders such as depression. We sought to investigate whether the number of children and presence of an infant predicted clinically significant sleep disturbance, and subsequently, depressive symptoms, for both mothers and fathers. We found that having an infant was significantly associated with increased wife WASO, which in turn was associated with increased depressive symptoms for wives. Further, husband WASO was associated with decreased wife WASO which was similarly associated with decreased depressive symptoms for wives. Thus, by helping with nighttime care so that mothers can get more sleep fathers can greatly contribute to the physical and emotional health of mothers.

Although findings from a pilot study indicate that caregivers of persons with dementia (PwD) report feeling stigmatized (Burgener, 2007; Burgener & Buckwalter, 2010), little research has been conducted on the effects of perceived stigma on fostering depressive symptoms among caregivers of PwD. The purpose of this study was to examine the relationship between perceived stigma and depressive symptoms among caregivers of PwD. The Modified Labeling Theory (MLT) developed by Link et. al. (1987; 1989) served as the organizing framework. The design of this study was a mixed methods approach including a descriptive longitudinal design with a qualitative interview. Caregivers of PwD (n=51) were interviewed regarding ethnic background, geographic location (rural and urban), knowledge of dementia, perceived stigma, depressive symptoms, and were asked to rate the extent of PwD's behavioral symptoms, while PwD (n=47) were assessed on their mental ability and disease stage. Caregivers were also asked to share their perceptions of stigma and mood change once their family members were diagnosed. Pearson product-moment correlations and a linear mixed model analysis determined the relationship between variables; for qualitative analysis, a directed approach to content analysis was applied. Findings indicated that caregivers' perceptions of stigma were significantly associated with depressive symptoms, both at baseline (r = 0.357, p = 0.0175) and over 18 months (p = 0.0045). Results also indicated that caregivers of PwD felt more depressed when they perceived additional stigma (p = 0.0019), regardless of caregiver ethnicity/race and caregivers' reactions in response to PwD memory and behavior problems. Moreover, perceived stigma minimally mediated the effect between caregivers' reactions toward the PwD's memory and behavior problems and depressive symptoms (14.4% decrease in the coefficient). Analysis of the qualitative data provided preliminary validation of the MLT and an in-depth understanding of caregivers' mood change since the diagnosis of their family member. Together our findings suggest that depressive symptoms among caregivers of PwD in response to the stresses of perceived stigma underscore the seriousness of this social problem. There is a need for effective interventions to combat caregivers' perceived stigma in order to enhance their psychological well-being.

BD is a severe and highly prevalent psychiatric illness characterized by oscillating mood episodes, where patients express either depressed mood, anhedonia, decreased activation along with concentration difficulties and sleep disturbances, or elevated mood with hyperactivity and loss of inhibitions. Between mood episodes, patients return to a relatively normal state of functioning without mood symptoms. Previous research on underlying neuronal mechanisms has led to a model of neuronal dysfunction in BD which states that BD arises from disruption in early development within brain networks that modulate emotional behavior. These abnormalities in the structure and function of key emotional control networks then lead to decreased connectivity among ventral prefrontal networks and limbic brain regions. This in turn creates a loss of emotional homeostasis, putting bipolar patients at risk for developing extreme mood states and switching among mood states. Two core components for BD have been identified, a hyperactive emotion processing system and a hypoactive cognitive functions system. It is controversial whether these deficits are still detectable in euthymia, so it is unclear if hyper- and hypoactivations represent state or trait-like characteristics. The aim of this study was to research both core components of BD with a paradigm eliciting differential activations in both cognitive and emotion processing networks. For this, an emotional word working memory paradigm was constructed to test for differences between manic, depressive, and remitted patients as well as a healthy control group. Differences were assessed in behavior, brain activation (as a correlate for the hypoactive cognitive functions system), measured with near-infrared spectroscopy (fNIRS), and electrophysiological changes in the late positive potential (as a correlate for the hyperactive emotion processing system), an event-related potential (ERP) measured with electroencephalography. 47 patients in the acutely ill phase and 45 healthy controls were measured. Of the 47 patients, 18 returned to the clinic for a second testing while in remission for at least 3 months. Acutely ill patients were classified into 4 groups according to their disorder status: a mildly depressed group, a depressed group, a manic group, and a mixed group along DSM-IV criteria. Analyses were calculated for 3 load conditions (1-back, 2-back and 3-back) and 3 valence conditions (negative, neutral, positive) for behavioral measures reaction time and omission errors, for brain activation and event related potential changes. Results indicate that ill patients differed from controls in their behavioral performance, but the difference in performance was modulated by the mood state they were in. Depressed patients showed the most severe differences in all behavioral measures, while manic and mixed patients differed from controls only upon different valence conditions. Brain activation changes were most pronounced in mildly depressed and manic patients, depressed patients and mixed patients did not differ as much from controls. ERP changes showed a significant difference only between mixed patients and controls, where mixed patients had an overall much higher ERP amplitude. When remitted patients were compared to controls, no differences in behavior, brain activation or ERP amplitude could be found. However, the same was true for differences in patients between acutely ill and remitted state. When looking at the overall data, the following conclusion can be drawn: assuming that the brain activation seen in the prefrontal cortex is part of the dorsal cognitive system, then this is the predominantly disturbed system in depressed patients who show only small changes in the ERP. In contrast, the predominantly disturbed system in manic and mixed patients is the ventral emotion processing system, which can be seen in a hyper-activation of ERP related neural correlates in mixed and hypo-activated neural correlates of the LPP in manic patients. When patients are remitted, the cognitive system regains temporary stability, and can be compared to that of healthy controls, while the emotion processing system remains dysfunctional and underlies still detectable performance deficits. / Die bipolare Störung ist eine schwere und hochprävalente psychiatrische Erkrankung, welche gekennzeichnet ist durch oszillierende Stimmungsepisoden, in denen Patienten entweder unter Anhedonie leiden, über Aktivitätsverlust und Konzentrationsstörungen klagen und Schlafstörungen haben, oder in deutlich aufgehellter Stimmung sind, hyperaktiv werden und soziale Hemmungen verlieren. Zwischen diesen Stimmungs-extremen durchlaufen die Patienten Phasen mit Stimmungsnormalisierung, oft ohne weitere schwere kognitive Defizite. Bisherige Studien über die zugrundeliegenden neuronalen Mechanismen haben ein Model hervorgebracht, welches von einer Störung der frühen Entwicklung in Hirnregionen, die emotionales Verhalten regulieren, ausgeht. Diese Anomalitäten in Struktur und Funktion von Kernkomponenten der Emotionskontrolle führen dann zu einem Verlust der Konnektivität in ventralen präfrontalen und limbischen Netzwerken. Dieser Verlust wiederum verursacht einen Verlust an emotionaler Homöostase, welches die Patienten dem Risiko aussetzt, extreme Stimmungsschwankungen zu erfahren. Zwei Kernkomponenten der bipolaren Störung wurden aufgrund dieses Modells definiert: ein hyperaktives Emotionsverarbeitungssystem, und ein hypoaktives kognitives Funktionssystem. Es ist bis jetzt nicht klar, in welcher Art und Weise diese emotionalen und kognitiven Dysfunktionen auch im euthymen Zustand weiterbestehen. Das Ziel dieser Studie war es, die beiden Kernkomponenten der Dysfunktion in der bipolaren Störung mit einem Paradigma zu untersuchen, welche beide Komponenten erfasst. Es wurde dazu ein emotionales Arbeitsgedächtnis Paradigma entwickelt, um Unterschiede zwischen akut kranken Patienten, gesunden Kontrollen und denselben Patienten im remittierten Zustand zu erfassen. Die Unterschiede sollten als Unterschiede der Reaktionszeit und Auslassungsfehler im Verhalten erfasst werden, ebenso als Unterschiede der Hirnaktivierung, gemessen mit funktionaler Nah-Infrarot Spektroskopie, und als Unterschiede in einem neurophysiologischen Korrelat, des „Late Positive Potential“ (LPP) betrachtet werden. 47 Patienten wurden rekrutiert, und eingeteilt nach dem Pol ihrer aktuellen Stimmungsepisode in schwer depressive Patienten, Patienten mit einer mittleren Depression, manische Patienten und Patienten im Mischzustand. Von den 47 akut kranken Patienten konnten 18 im remittierten Zustand wiederum gemessen werden. Anschließend wurden Gruppenunterschiede in 3 kognitiven Variablen (1-back, 2-back und 3-back) und 3 emotionalen Variablen (positiv, neutral, negativ) für Verhalten, Hirnaktivierung und Amplitudenänderung in der LPP berechnet. Die Ergebnisse zeigen dass akut kranke Patienten sich in ihrem Verhalten von Kontrollen unterscheiden, jedoch wurden diese Unterschiede von der Art der aktuellen Stimmungsepisode moduliert. Schwer depressive Patienten zeigten die deutlichsten Unterschiede, während manische Patienten und Patienten im Mischzustand nur in den emotionalen Variablen Unterschiede zeigten. Die Hirnaktivierungsunterschiede waren am deutlichsten zwischen Patienten mit einer mittelschweren Depression und manischen Patienten, bei schwer depressiven Patienten und Patienten im Mischzustand waren diese Unterschiede deutlich schwächer ausgeprägt. Die LPP Analysen zeigten deutliche Unterschiede nur zwischen Patienten mit Mischbild und Kontrollen, die Patienten hatten hierbei eine deutlich erhöhte LPP Amplitude. Die Untersuchung der Unterschiede zwischen remittierten Patienten und Kontrollen ergab keine signifikanten Ergebnisse, ebenso die Analysen der Unterschiede zwischen akut kranken und remittierten Patienten. Alle Ergebnisse zusammengenommen, ergibt sich folgendes Bild: Wenn die Hirnaktivierung als Korrelat eines gestörten kognitiven Systems gesehen werden kann, und die LPP als Korrelat eines gestörten Emotionsverarbeitungssystems, dann könnte für Patienten mit einer mittleren oder schweren Depression das kognitive System das Hauptproblem darstellen, während für manische Patienten und Patienten im Mischzustand das Emotionsverarbeitungssystem das dominante Problem darstellt. Wenn die Patienten dann remittieren, erhält das kognitive System eine vorübergehende Stabilität zurück, das Emotionsverarbeitungssystem jedoch bleibt dysfunktional, und ist verantwortlich für die bestehenden emotionalen und kognitiven Defizite.

Die bipolare Störung ist eine psychische Erkrankung, die sich durch wiederkehrende depressive und (hypo-) manische Phasen auszeichnet. Neben Stimmungsschwankungen leiden viele Patienten unter kognitiven Beeinträchtigungen, die nicht nur während akuter Episoden, sondern auch in der Remission, d.h. in euthymer Stimmungslage persistieren. Die vorliegende Arbeit beschäftigte sich mit den klinischen Korrelaten von kognitiven Defiziten und der Effektivität eines kognitiven Trainings bei bipolaren Patienten (BP). In der ersten Teilstudie wurde untersucht, wie sich die kognitive Leistung der Patienten von der akuten Phase bis zur Remission verändert. Dazu wurden 55 akut depressive und (hypo-) manische BP und 55 gesunde Kontrollpersonen wiederholt mit einer neuropsychologischen Testbatterie untersucht. 29 Patienten konnten nach mindestens 3-monatiger Remission erneut getestet werden. Die Ergebnisse zeigen, dass die akut kranken BP domänenübergreifend kognitive Störungen im Vergleich zu gesunden Kontrollen aufweisen, wobei die depressiven Patienten eher in der Verarbeitungsgeschwindigkeit, der Aufmerksamkeit und dem Gedächtnis beeinträchtigt waren. Die akut manischen Patienten hatten hingegen auffällige Defizite in den exekutiven Funktionen. Die Performanz der BP besserte sich zwar in der Remission, es waren aber weiterhin im Vergleich zu den Kontrollen Defizite in der psychomotorischen Geschwindigkeit, dem Arbeitsgedächtnis und dem verbalen Gedächtnis festzustellen. Es zeigte sich außerdem, dass die Verarbeitungsgeschwindigkeit, die Aufmerksamkeit und das verbale Gedächtnis in Zusammenhang mit subdepressiven Symptomen und Schlafstörungen standen, wohingegen die exekutiven Testmaße nicht mit diesen „State“-Faktoren assoziiert waren. Diese Ergebnisse lassen vermuten, dass die exekutiven Funktionen als Trait-Merkmale der bipolaren Störung in Frage kommen, wohingegen Aufmerksamkeit und Gedächtnis durch das Vorliegen von Residualsymptomen beeinträchtigt sind. Ziel des zweiten Teils dieser Arbeit war es, eine kognitive Defizit- vs. Nondefizit Subgruppe innerhalb der BP zu identifizieren, um herauszufinden welche soziodemographischen oder krankheitsrelevanten Charakteristika mit kognitiven Störungen in Zusammenhang stehen. Dazu wurde die neuropsychologische Testleistung von 79 euthymen BP und 70 gesunden Kontrollen verglichen. Es zeigte sich erwartungsgemäß, dass die BP in der psychomotorischen Geschwindigkeit, der Aufmerksamkeit, dem Arbeitsgedächtnis, dem verbalen Gedächtnis, der Wortflüssigkeit und dem problemlösenden Denken trotz stabiler Remission signifikant schlechtere Leistungen erbrachten als die gesunden Kontrollen. Im Anschluss wurde die bipolare Stichprobe anhand ihrer Testleistung in eine Defizit- und eine Nondefizit Gruppe aufgeteilt. Die Ergebnisse zeigen, dass 54% der BP in allen Tests eine völlig normgerechte Leistung erbrachten. Die Studie bestätigte demnach, dass nicht alle Patienten kognitive Defizite aufweisen, sondern Subgruppen bestehen, die sich in verschiedenen Variablen voneinander unterscheiden: Die Defizit-Subgruppe berichtete signifikant mehr subdepressive Symptome und es lagen häufiger persistierenden Schlafstörungen und die Diagnose einer komorbiden Erkrankung vor (Angststörung, ADHS und Migräne). Zudem zeigte sich ein Zusammenhang zwischen Polypharmazie und kognitiven Defiziten. Diese Ergebnisse demonstrieren, dass ein Teil der kognitiven Störungen bei BP durch eine nicht vollständige Remission und sekundäre Symptome bedingt sind. Es ergab sich keine Assoziation zwischen kognitiver Leistung und krankheitsrelevanten Variablen, wie z.B. Anzahl der Phasen, Bipolar-Subtyp oder Ersterkrankungsalter. Diese Daten widersprechen zwar nicht der Hypothese, dass kognitive Störungen durch neurodegenerative Prozesse bedingt sind, sie weisen jedoch darauf hin, dass bei der bipolaren Störung häufig Residualsymptome vorliegen, welche im Rahmen von Studie als auch bei der therapeutischen Arbeit stärker als bisher berücksichtigt werden müssen. In beiden Teilstudien zeigte sich zudem, dass kognitive Störungen mit einem reduzierten psychosozialen Funktionsniveaus in Verbindung stehen. Dieses Ergebnis steht in Einklang mit bisherigen Untersuchungen, die berichten, dass Patienten mit kognitiven Defiziten soziale und berufliche Einschränkungen aufweisen, die wiederum mit einem schlechteren Krankheitsverlauf assoziiert ist. Aufgrund dessen wurde von einigen Autoren vorgeschlagen, mit Hilfe spezieller Interventionen wie der kognitiven Remediation (KR) die geistigen Funktionen zu rehabilitieren. In der vorliegenden Interventionsstudie wurde deshalb der Frage nachgegangen, ob die neurokognitive Leistungsfähigkeit und das psychosoziale Funktionsniveau der bipolaren Stichprobe durch KR verbessert werden kann. Zudem sollte untersucht werden, inwiefern kognitives Training zu Veränderungen der präfrontalen Hirnaktivität führt. Dafür wurde vor und nach dem Training eine Messung mit der Methode der funktionellen Nahinfrarotspektroskopie (fNIRS) durchgeführt. Das 3-monatige KR-Programm bestand aus einem computerisierten kognitiven Training und der Vermittlung von kognitiven Skills im Rahmen von 12-wöchentlichen Gruppensitzungen. Im Anschluss an das Training wurden die Teilnehmer (26 bipolare und als Vergleichsgruppe 13 unipolare Patienten) im Rahmen einer Post-Messung wiederholt untersucht. Zudem wurde zum Vergleich eine Kontrollgruppe von 10 BP im Abstand von 3 Monaten untersucht, die keine Intervention, sondern die Standardbehandlung erhielt. Aufgrund zahlreicher Drop-Outs konnten am Ende des Erhebungszeitraums die Daten von 16 bipolaren und 10 unipolar depressiven Patienten ausgewertet werden. Die Trainingsteilnehmer erbrachten im Gegensatz zu der Kontrollgruppe signifikante Leistungssteigerungen in den Tests zur Erfassung der psychomotorischen Geschwindigkeit, dem Arbeitsgedächtnisses, dem verbalen Gedächtnis und dem problemlösenden Denken. Zudem zeigte sich nach dem Training eine Verbesserung des psychosozialen Funktionsniveaus und eine Reduktion der subdepressiven Symptomatik. Eine Veränderung der präfrontalen Hirnaktivierung konnte jedoch nicht verifiziert werden. Die Ergebnisse lassen demnach schlussfolgern, dass Patienten mit affektiven Störungen von einem kognitiven Training profitieren, wobei die damit einhergehenden funktionalen Veränderungen der Hirnaktivität in Studien mit größeren Stichproben untersucht werden müssen. / Bipolar disorder is a recurrent and highly disabling affective disorder characterized by mood instability as well as cognitive disturbances. Considerable evidence has demonstrated that neuropsychological deficits are prevalent both during acute episodes and euthymia. Recent studies verified cognitive impairments during euthymia with deficits up to one standard deviation below average in psychomotor speed, attention, working memory, long term memory and executive functioning. Up to now, the mechanisms behind why some bipolar patients (BP) do not reach their former level of cognitive performance and psychosocial functioning while others remit completely, are not understood. In the first part of this doctoral thesis a longitudinal study was conducted which aimed to characterize changes in cognitive functioning from acute illness to remission in BP. For this purpose, 55 acutely admitted BP (35 depressed, 20 hypo-/ manic) and 55 healthy controls (HC) were tested with an extended neuropsychological test battery (attention, working memory, verbal memory and executive functioning). The measurement was repeated after three month of remission in 29 BP and 55 controls. The results showed global impairments in acutely admitted BP compared to HC. In particular, depressed patients showed a characteristic psychomotor slowing while (hypo-) manic patients had severe deficits in working memory and frontal-executive functions. During remission, cognitive functions in BP partially recovered while psychomotor speed, working memory and verbal memory were still impaired. Furthermore, it was shown that “state” factors such as subclinical symptoms and persisting sleep disorder were significantly associated with speed, attention and verbal memory. In contrast, working memory was correlated with psychotic symptoms (life time), indicating this cognitive domain as trait-related. Therefore, the results of this study indicate that residual symptoms could be potential confounders in studies on cognition and the search for cognitive endophenotypes in BD. However, to clarify trait vs. disease process effects in cognition in BD, more studies with a prospective design are needed. The second study of the present thesis aimed to identify a “cognitive deficit vs. non-deficit” subgroup within BD to find clinical correlates of cognitive impairments. Therefore, the neuropsychological test performance of 79 euthymic outpatients (BD-I and II) was compared to 70 matched, healthy controls. As expected, the sample of euthymic BP performed significantly worse than controls in psychomotor speed, divided attention, working memory, verbal memory, word fluency and problem solving. However, 54 % of the patients did not have any neurocognitive deficits at all, and whether or not a patient belonged to the non-deficit group was not influenced by disease severity (e.g. number of previous episodes, duration of illness, age of onset or bipolar subtype). Instead, our results demonstrated that patients suffering from persistent sleep disturbances and sub-threshold depressive symptomatology show more severe cognitive dysfunctions. In addition, comorbid disorders like anxiety, ADHD and migraine were associated with cognitive deficits. Furthermore, polypharmacy seemed to have a negative influence on cognitive functioning in BP. In sum, these results suggest that a major part of cognitive impairment in bipolar disorder is due to partial remission and secondary symptoms, especially sleep disorder and sub-syndromal depression. Even though our results do not disprove the idea of neurodegenerative processes in bipolar disorder, our data underscore the importance of residual symptoms. This has implications for future research and calls for the improvement of treatment options for cognitively impaired BP. The results of both studies reported above, showed an association between cognitive deficits and low psychosocial functioning. This is in line with previous studies showing strong associations between cognitive impairment and low social or occupational functioning which has in turn negative effects on illness progression. Therefore, many researchers suggest interventions not only to improve affective symptoms, but also cognitive dysfunctions. Cognitive remediation (CR) is a new psychological treatment specifically targeting cognitive functioning and coping skills, which is commonly assessed in patients with schizophrenia. Although BP have less severe, but similar cognitive deficits compared to patients suffering from schizophrenia, there is only little evidence on the efficacy of CR programs in patients with affective disorders. In the present thesis we developed a tailored CR program to evaluate the effect on the patients’ objective neuropsychological performance and everyday psychosocial functioning. Furthermore, we assessed the effect of CR on neural activity. For this purpose we applied functional near-infrared spectroscopy (fNIRS) to measure frontal cortex activation prior and after CR. The CR program was composed of 12 group sessions which were conducted weekly. It consisted of a computerized training of several cognitive functions and a cognitive-skills part including mindfulness-based strategies as well as memorizing, problem solving, verbal reasoning, planning, organization in daily life and effective communication techniques. 39 bipolar und unipolar depressive patients took part in the cognitive training, however, due to several drop-outs only 26 patients could be measured again after the training. A comparable control group consisting of 10 BP was measured at baseline and follow-up after three month of treatment. Within the patient group a significant improvement of cognitive performance after cognitive training could be assessed: Compared to the control group, BP showed an improved performance in divided attention, working memory, verbal memory and problem solving after the CR program; unipolar depressive patients additionally improved in psychomotor speed and susceptibility to interference. Both bipolar and unipolar patients showed a higher psychosocial functioning and a reduction of subdepressive symptoms after the CR program. However, we did not reveal any pre-post changes in neural activity of the prefrontal cortex as measured by fNIRS. Therefore, our results demonstrate that patients with affective disorders benefit from CR with regard to mood symptoms and neurocognitive/ psychosocial functioning. However, the relation between behavioral changes and changes in functional activation following CR requires further studies.