In situations of real threat, showing a fear reaction makes sense, thus, increasing the chance to survive. The question is, how could anybody differentiate between a real and an apparent threat? Here, the slogan counts “better safe than sorry”, meaning that it is better to shy away once too often from nothing than once too little from a real threat. Furthermore, in a complex environment it is adaptive to generalize from one threatening situation or stimulus to another similar situation/stimulus. But, the danger hereby is to generalize in a maladaptive manner involving as it is to strong and/or fear too often “harmless” (safety) situations/stimuli, as it is known to be a criterion of anxiety disorders (AD). Fear conditioning and fear generalization paradigms are well suited to investigate fear learning processes. It is remarkable that despite increasing interest in this topic there is only little research on fear generalization. Especially, most research on human fear conditioning and its generalization has focused on adults, whereas only little is known about these processes in children, even though AD is typically developing during childhood. To address this knowledge gap, four experiments were conducted, in which a discriminative fear conditioning and generalization paradigm was used. In the first two experiments, developmental aspects of fear learning and generalization were of special interest. Therefore, in the first experiment 267 children and 285 adults were compared in the differential fear conditioning paradigm and generalization test. Skin conductance responses (SCRs) and ratings of valence and arousal were obtained to indicate fear learning. Both groups displayed robust and similar differential conditioning on subjective and physiological levels. However, children showed heightened fear generalization compared to adults as indexed by higher arousal ratings and SCRs to the generalization stimuli. Results indicate overgeneralization of conditioned fear as a developmental correlate of fear learning. The developmental change from a shallow to a steeper generalization gradient is likely related to the maturation of brain structures that modulate efficient discrimination between threatening and (ambiguous) safety cues. The question hereby is, at which developmental stage fear generalization gradients of children adapt to the gradients of adults. Following up on this question, in a second experiment, developmental changes in fear conditioning and fear generalization between children and adolescents were investigated. According to experiment 1 and previous studies in children, which showed changes in fear learning with increasing age, it was assumed that older children were better at discriminating threat and safety stimuli. Therefore, 396 healthy participants (aged 8 to 12 years) were examined with the fear conditioning and generalization paradigm. Again, ratings of valence, arousal, and SCRs were obtained. SCRs indicated differences in fear generalization with best fear discrimination in 12-year-old children suggesting that the age of 12 years seems to play an important role, since generalization gradients were similar to that of adults. These age differences were seen in boys and girls, but best discrimination was found in 12-year-old boys, indicating different development of generalization gradients according to sex. This result fits nicely with the fact that the prevalence of AD is higher in women than in men. In a third study, it was supposed that the developmental trajectory from increased trait anxiety in childhood to manifest AD could be mediated by abnormal fear conditioning and generalization processes. To this end, 394 children aged 8 to 12 years with different scores in trait anxiety were compared with each other. Results provided evidence that children with high trait anxiety showed stronger responses to threat cues and impaired safety signal learning contingent on awareness as indicated by arousal at acquisition. Furthermore, analyses revealed that children with high trait anxiety showed overall higher arousal ratings at generalization. Contrary to what was expected, high trait anxious children did not show significantly more fear generalization than children with low trait anxiety. However, high-trait-anxious (HA) participants showed a trend for a more linear gradient, whereas moderate-trait-anxious (MA) and low-trait-anxious (LA) participants showed more quadratic gradients according to arousal. Additionally, after controlling for age, sex and negative life experience, SCR to the safety stimulus predicted the trait anxiety level of children suggesting that impaired safety signal learning may be a risk factor for the development of AD. Results provide hints that frontal maturation could develop differently according to trait anxiety resulting in different stimuli discrimination. Thus, in a fourth experiment, 40 typically developing volunteers aged 10 to 18 years were screened for trait anxiety and investigated with the differential fear conditioning and generalization paradigm in the scanner. Functional magnetic resonance imaging (fMRI) were used to identify the neural mechanisms of fear learning and fear generalization investigating differences in this neural mechanism according to trait anxiety, developmental aspects and sex. At acquisition, HA participants showed reduced activation in frontal brain regions, but at generalization, HA participants showed an increase in these frontal regions with stronger linear increase in activation with similarity to CS+ in HA when compared to LA participants. This indicates that there is a hyper-regulation in adolescents to compensate the higher difficulties at generalization in form of a compensatory mechanism, which decompensates with adulthood and/or may be collapsed in manifest AD. Additionally, significant developmental effects were found: the older the subjects the stronger the hippocampus and frontal activation with resemblance to CS+, which could explain the overgeneralization of younger children. Furthermore, there were differences according to sex: males showed stronger activation with resemblance to CS+ in the hippocampus and frontal regions when compared to females fitting again nicely with the observation that prevalence rates for AD are higher for females than males. In sum, the studies suggest that investigating developmental aspects of (maladaptive) overgeneralization may lead to better understanding of the mechanisms of manifest anxiety disorders, which could result in development and provision of prevention strategies. Although, there is need for further investigations, the present work gives some first hints for such approaches.

Angsterkrankungen stellen mit einer 12-Monats-Prävalenz von 14% die häufigsten psychischen Erkrankungen in der westlichen Gesellschaft dar. Angesichts der hohen querschnittlichen wie sequentiellen Komorbidität von Angsterkrankungen, der ausgeprägten individuellen Einschränkungen sowie der hohen ökonomischen Belastung für das Gesundheitssystem ist neben therapeutischen Behandlungsansätzen die Entwicklung von kurzzeitigen, kostengünstigen und leicht zugänglichen Präventionsmaßnahmen von großer Bedeutung und steht zunehmend im Fokus des gesundheitspolitischen Interesses, um die Inzidenz von Angsterkrankungen zu reduzieren. Voraussetzung für die Entwicklung von gezielten und damit den effektivsten Präventionsmaßnahmen sind valide Risikofaktoren, die die Entstehung von Angsterkrankungen begünstigen. Ein Konstrukt, das in der Literatur als subklinisches Symptom in Form einer kognitiven Vulnerabilität für Angsterkrankungen und damit als Risikofaktor angesehen wird, ist die sogenannte Angstsensitivität (AS). AS umfasst die individuelle Tendenz, angstbezogene körperliche Symptome generell als bedrohlich einzustufen und mit aversiven Konsequenzen zu assoziieren. Das Ziel der vorliegenden Arbeit war daher die Etablierung und Validierung eines Präventionsprogramms zur Reduktion der AS an einer nicht-klinischen Stichprobe von 100 Probanden (18-30 Jahre) mit einer erhöhten AS (Anxiety Sensitivity Index [ASI-3] ≥17) sowie die Rekrutierung von 100 alters- und geschlechtsangeglichenen Probanden mit niedriger Angstsensitivität (ASI-3 <17). In einem randomisiert-kontrollierten Studiendesign durchliefen die Probanden mit hoher AS entweder das über fünf Wochen angelegte „Kognitive Angstsensitivitätstraining“ (KAST) als erste deutschsprachige Übersetzung des Computer-basierten „Cognitive Anxiety Sensitivity Treatment“ (CAST) von Schmidt et al. (2014) oder wurden der Wartelisten-Kontrollgruppe zugeteilt. Das KAST Training bestand aus einer einmaligen Vermittlung kognitiv-behavioraler Psychoedukation zum Thema Stress und Anspannung sowie deren Auswirkungen auf den Körper und der Anleitung von zwei interozeptiven Expositionsübungen (‚Strohhalm-Atmung‘ und ‚Hyperventilation‘), die über den anschließenden Zeitraum von fünf Wochen in Form von Hausaufgaben wiederholt wurden. Es konnte gezeigt werden, dass die Teilnehmer des KAST-Programms nach Beendigung des Trainings (T1) eine signifikant niedrigere AS-Ausprägung im Vergleich zur Wartelisten-Kontrollgruppe aufwiesen und diese Reduktion auch über den Katamnese-Zeitraum von sechs Monaten (T2) stabil blieb. Ergänzend wurde auch die Targetierbarkeit weiterer intermediärer Risikomarker wie der Trennungsangst (TA), des Index der kardialen Sensitivität sowie der Herzratenvariabilität (HRV) untersucht, die jedoch nicht durch das KAST-Training direkt verändert werden konnten. Im Vergleich der Subgruppen von Probanden mit hoher AS und gleichzeitig hoher TA (Adult Separation Anxiety Questionnaire [ASA-27] ≥22) und Probanden mit hoher AS, aber niedriger TA (ASA-27 <22) zeigte sich, dass die AS-TA-Hochrisikogruppe ebenfalls gut von der KAST-Intervention profitieren und eine signifikante Reduktion der AS erzielen konnte, indem sie sich bei T1 dem Niveau der Gruppe mit niedriger TA anglich. Zudem korrelierte die prozentuale Veränderung der Einstiegswerte der inneren Anspannung während der Strohhalm-Atmungsübung positiv mit der prozentualen Veränderung der dimensionalen TA bei T1. Zusammenfassend weisen die Ergebnisse der vorliegenden Arbeit erstmalig auf die Wirksamkeit der deutschsprachigen Übersetzung des CAST-Programms (Schmidt et al., 2014), eines Computer-basierten, und damit leicht zu implementierenden sowie kostengünstigen Programms, in Bezug auf die Reduktion der AS sowie indirekt der TA hin und können damit zur indizierten und demnach besonders effektiven Prävention von Angsterkrankungen in Hochrisikogruppen beitragen.