Der Persönlichkeitsfaktor Impulsivität und eine verminderte Inhibitionsfähigkeit werden als Risikofaktoren für Suchterkrankungen diskutiert. Impulsivität ist ein multidimensionales Konstrukt und lässt sich durch Fragebögen wie die Barratt Impulsiveness Scale (BIS-11) quantifizieren. Inhibitionsfähigkeit kann mit Hilfe eines Stop-Signal Task operationalisiert werden und wird von einigen Autoren als Subdomäne von Impulsivität angesehen (Response Inhibition). Welche Domänen des Persönlichkeitsmerkmals Impulsivität die behavioralen und neuronalen Korrelate von Inhibitionsfähigkeit abbilden, ist allerdings bis jetzt nicht eindeutig geklärt. Auch Veränderungen der Belohnungswahrnehmung tragen zur Pathogenese von Suchterkrankungen bei, und belohnungsassoziierte Aktivierungen im ventralen Striatum weisen bei Gesunden eine positive Beziehung zu Impulsivität auf. Die impulsivitätsabhängigen Auswirkungen von Belohnungsverarbeitung auf die Inhibitionsfähigkeit sind jedoch unzureichend verstanden. Zur weiteren Untersuchung der Beziehung von Impulsivität und Inhibitionsfähigkeit sowie ihrer Interaktion mit Belohnungsverarbeitung selektierten wir aus einer Gruppe von 452 Probanden je 26 hoch und niedrig impulsive Probanden, deren durchschnittliche BIS-11 Werte den Kriterien für hohe bzw. niedrige Impulsivität entsprachen. Alle 52 Probanden absolvierten einen Stop-Signal Task mit integrierter Belohnungskomponente während einer simultanen funktionellen MRT Messung (fMRT). Neben einer neuropsychologischen Testung erhielten die Probanden eine umfassende Charakterisierung ihrer Persönlichkeits- und Impulsivitätsdomänen durch die Fragebögen UPPS (Urgency, Lack of Premeditation, Lack of Perseverance, Sensation Seeking), Sensation Seeking Scale und NEO-Fünf-Faktoren-Inventar. Die Auswertung der Verhaltensdaten erfolgte mit SPSS, die Analyse der fMRT-Daten nach einer Vorverarbeitung mit SPM getreu des General Linear Model. Für die Extremgruppen des BIS-11 fanden wir, entgegen unserer anfänglichen Vermutung, keinen signifikanten Gruppenunterschied, weder hinsichtlich der behavioral gemessenen Inhibitionsfähigkeit (SSRT), noch im neuronalen Aktivierungsmuster des Inhibitionsnetzwerks. Eine feinere Untersuchung unter Einbezug der Impulsivitätsubdomänen zeigte, dass die Subdimension Urgency die individuelle Inhibitionsfähigkeit der Probanden am besten vorhersagte und eine positive Korrelation mit dem Inhibitionsmaß SSRT aufwies. Interessanterweise zeigten Urgency-Werte zusätzlich auch eine negative Korrelation mit den neuronalen Aktivierungen im rechten inferior-frontalen Gyrus, einer Schlüsselregion des Inhibitionsnetzwerks. Belohnte erfolgreiche Stop-Trials wiesen gegenüber nicht erfolgreichen belohnten Stop-Trials auf neuronaler Ebene eine verstärkte Aktivierung des ventralen Striatums auf. Diese Aktivierung zeigte Ähnlichkeit mit einem Prädiktionsfehlersignal und könnte möglicherweise ein selbst generiertes Feedbacksignal darstellen. Nur Probanden mit niedrigen Urgency-Werten konnten dieses potentielle Feedbacksignal für eine verbesserte Inhibition (kürzere SSRT) nutzen, während sich dieser Zusammenhang nicht für Probanden mit höheren Urgency-Werten zeigte. Die Ergebnisse verdeutlichen, dass die Beziehung zwischen Impulsivität und behavioraler Inhibitionsfähigkeit vorsichtig und gezielt multidimensional betrachtet werden sollte. Während der Gesamtscore des BIS-11 Response Inhibition nicht ausreichend abzubilden scheint, zeigte die Subdimension Urgency sowohl behavioral als auch neuronal eine Assoziation mit einer verminderten Inhibitionsfähigkeit. Interessanterweise weist Urgency auch eine Verbindung zur Belohnungsverarbeitung im Stop-Signal Task auf und scheint mit dem Nutzen von potentiellen individuellen Feedbacksignalen verknüpft zu sein. Da Urgency-Werte sich bei verschiedenen psychischen Störungen, wie Abhängigkeitserkrankungen, erhöht zeigen, stellt Urgency möglicherweise ein Bindeglied zwischen Response Inhibition und Belohnungsverarbeitung dar. Aus diesen Gründen sollte Urgency in zukünftigen Studien zu den Entstehungsmechanismen von Suchterkrankungen als Moderator von behavioraler Handlungsadaption miteinbezogen werden.

Happiness, anger, surprise, irritation… if we note down the emotions that we go through on a given day, the list will most probably be quite long. A surge of studies on the bidirectional interaction between emotion and cognition suggests that we need emotional appraisals in order to lead a successful life and maintain our personal, social and economic integrity (Bechara, 2005; Damasio, 1994; Fox, 2008; Gross & Thompson, 2007; Walter, 2005). And yet, we seldom ‘just’ experience emotions, but often try to influence them to best fit our current goals. Based on the assumption that emotional reactions entail changes on various levels, and that these changes happen in- or outside of our awareness, affective science has adopted emotion regulation as one of its major research topics (Beauregard, Levesque, & Paquette, 2004; Gross, 1999; Ochsner, 2007). In fact, neural (e.g. amygdala activation) and behavioral (e.g. feeling of negativity) correlates of emotional reactions are effectively reduced by top-down processes of explicit and implicit control (Drabant, McRae, Manuck, Hariri, & Gross, 2009; Levesque, et al., 2003; Ochsner, Ray, et al., 2004). Furthermore, evidence from studies investigating voluntary thought control suggests that control strategies may have lasting and paradoxical consequences (Abramowitz, Tolin, & Street, 2001; Wegner, 2009). In a very recent investigation, lasting effects of regulation were also shown after the cognitive control of emotions: the activation timecourse of the amygdala was significantly increased immediately following regulation, and this difference was also related to the activation of the amygdala to the same stimuli a few minutes later (Walter, et al., 2009). Aside from these contextual or qualitative influences, emotional processing also differs between individuals: genetic variation within the serotonergic system for instance is known to affect emotional reactivity both on the behavioral and on the neural level (Hariri, et al., 2005; Hariri, et al., 2002; Lesch, et al., 1996). In the present work, the temporal dynamics of volitional emotion regulation were investigated in three studies. It was hypothesized that both the subjective experience of negativity and the amygdala activation can be attenuated by the detachment from negative emotions, which in turn leads to an immediate neural aftereffect after the offset of regulation. Furthermore, volitional emotion regulation was expected to be capable of reducing or even obliterating genetically mediated amygdala hyperreactivity to negative emotional cues. Similar to previous investigations (Walter, et al., 2009), pictures of aversive or neutral emotional content were presented while participants were instructed to react naturally to half of the pictures, and to regulate their emotional response upon the other half of the stimuli. The first two studies of the present work were designed to further characterize the immediate aftereffect of volitional regulation in the amygdala: Study 1 included behavioral ratings of negativity at picture offset and at fixation offset in order to provide behavioral measures of experiential changes, while in Study 2, participants continued to experience or regulate their emotions during a “maintain” phase after picture offset. The primary goal of Study 3 was to evaluate whether volitional emotion regulation can reduce genetically mediated amygdala hyperreactivity to aversive emotional material in individuals with the short variant of the serotonin transporter genotype (Hariri, et al., 2005; Hariri, et al., 2002), and whether the immediate aftereffect is also influenced by the serotonin transporter genotype. In all three studies, the amygdala was significantly activated by aversive versus neutral stimuli, while cognitive emotion regulation attenuated the activation in the amygdala and increased the activation in a frontal-parietal network of regulatory brain regions. This neural effect was complemented by the behavioral ratings which show that the subjective experience of negativity was also reduced by detachment (Study 1). Also in all three studies, an immediate aftereffect was observed in the amygdala following the end of regulation. Moreover, the preoccupation with the previously seen pictures after the scanning session varied across the experimental conditions (Studies 2 and 3). Volitional regulation proved effective in reducing amygdala activation to negative stimuli even in 5-HTTLPR short allele carriers that show an increased reactivity to this type of cue. At the same time, functional coupling of the ventrolateral and medial orbital prefrontal cortex, the subgenual and the rostral anterior cingulate with the amygdala was higher in the s-group. However, in Study 3 the immediate aftereffect was found only in l/l-homozygote individuals following the regulation of fear. Taken together, the results of the three studies clearly show that volitional regulation is effective in reducing behavioral and neural correlates of the experience of negative emotions (Levesque, et al., 2003; Ochsner, Bunge, Gross, & Gabrieli, 2002; Ochsner, Ray, et al., 2004), even in the case of a genetically mediated hyperreactivity to such materials. Thus, it seems reasonable to assume that conscious will can effectively counteract genetic determinants of emotional behavior. Moreover, the present results suggest that the temporal dynamics of volitional emotion regulation are characterized by a paradoxical rebound in amygdala activation after regulation, and that the immediate aftereffect is a marker of the efficiency of the initial and the sustained effects of emotion regulation (Walter, et al., 2009). In summary, the successful replication of the immediate aftereffect of emotion regulation in all three studies of this dissertation opens up exciting new research perspectives: a comparison of the short- and long-term effects of different regulatory strategies, and the investigation of these effects also in positive emotions would complement the present results, since the neural mechanisms involved in these processes show some characteristic differences (Ochsner, 2007; Staudinger, Erk, Abler, & Walter, 2009). A comprehensive characterization of this neural marker and its implications for emotional experience might also be useful with respect to clinical applications. The detailed examination of the various time scales of emotional regulation might for instance inform the diagnostic and therapeutic interventions in affective disorders that are associated with emotional dysfunctions (Brewin, Andrews, & Rose, 2000; Johnstone, van Reekum, Urry, Kalin, & Davidson, 2007). Ultimately, we might thus come to understand the neural underpinnings of what the feelings we have today have to do with the feelings we had yesterday – and with the feelings with might have tomorrow.

Subjects participating in magnetic resonance imaging (MRI) examinations regularly report anxiety and stress related reactions. This may result in impaired data quality and premature termination of scans. Moreover, cognitive functions and neural substrates can be altered by stress. While prior studies investigated pre–post scan differences in stress reactions only, the present study provides an in-depth analysis of mood changes and hormonal fluctuations during the time course of a typical fMRI session. Thirty-nine subjects participated in the study. Subjective mood, salivary alpha-amylase (sAA) and cortisol were assessed at six time points during the lab visit. Associations between hormonal data and neural correlates of a visual detection task were observed using a region of interest approach applied to the thalamic region. Mood and hormonal levels changed significantly during the experiment. Subjects were most nervous immediately after entering the scanner. SAA was significantly elevated after MRI preparation. A subgroup of n = 5 (12.8%) subjects showed pronounced cortisol responses exceeding 2.5 nmol/l. Preliminary fMRI data revealed an association between sAA levels and left thalamic activity during the first half of the experiment that disappeared during the second half. No significant correlation between cortisol and thalamic activity was observed. Results indicate that an fMRI experiment may elicit subjective and neuroendocrine stress reactions that can influence functional activation patterns.

Although functional magnetic resonance imaging (fMRI) has gained increasing importance in investigating neural substrates of anxiety disorders, less is known about the stress eliciting properties of the scanner environment itself. The aim of the study was to investigate feasibility, self-reported distress and anxiety management strategies during an fMRI experiment in a comprehensive sample of patients with panic disorder and agoraphobia (PD/AG). Within the national research network PANIC-NET, n = 89 patients and n = 90 controls participated in a multicenter fMRI study. Subjects completed a retrospective questionnaire on self-reported distress, including a habituation profile and exploratory questions about helpful strategies. Drop-out rates and fMRI quality parameters were employed as markers of study feasibility. Different anxiety measures were used to identify patients particularly vulnerable to increased scanner anxiety and impaired data quality. Three (3.5%) patients terminated the session prematurely. While drop-out rates were comparable for patients and controls, data quality was moderately impaired in patients. Distress was significantly elevated in patients compared to controls; claustrophobic anxiety was furthermore associated with pronounced distress and lower fMRI data quality in patients. Patients reported helpful strategies, including motivational factors and cognitive coping strategies. The feasibility of large-scale fMRI studies on PD/AG patients could be proved. Study designs should nevertheless acknowledge that the MRI setting may enhance stress reactions. Future studies are needed to investigate the relationship between self-reported distress and fMRI data in patient groups that are subject to neuroimaging research.

Accumulating evidence suggests that the magnetic resonance imaging (MRI) scanner can act as a stressor, eliciting subjective and neuroendocrine stress responses. Approaches to familiarize subjects with the scanner could help minimizing unintended effects on neural activation patterns of interest. Controlled studies on the effects of a scanner training are however missing. Using a comparative design, we analyzed within- and between session changes in subjective and neuroendocrine stress parameters in 63 healthy, scanner-naïve adults who participated in a two-day training protocol in an MRI, mock, or lab environment. A habituation task was used to assess within-session changes in subjective and neuroendocrine (cortisol) stress parameters; between-session changes were indicated by differences between days. MRI and mock, but not lab training were successful in reducing subjective distress towards the scanner. In contrast, cortisol reactivity towards the training environment generally increased during day 2, and the percentage of cortisol responders particularly rose in the mock and MRI groups. Within-session habituation of subjective arousal and anxiety was observed during both days and irrespective of training condition. Present findings demonstrate that training in a scanner environment successfully reduces subjective distress, but may also induce sensitization of endocrine stress levels during repeated scanning. Subjective distress can further be stabilized by acclimating subjects to the environment prior to the MRI assessment, including a short habituation phase into the assessment protocol. If replicated, present findings should be considered by researchers employing repeated measurement designs where subjects are exposed to a scanner more than once.

Background: Learning by conditioning is a key ability of animals and humans for acquiring novel behavior necessary for survival in a changing environment. Aberrant conditioning has been considered a crucial factor in the etiology and maintenance of panic disorder with agoraphobia (PD/A). Cognitive-behavioral therapy (CBT) is an effective treatment for PD/A. However, the neural mechanisms underlying the effects of CBT on conditioning processes in PD/A are unknown. Methods: In a randomized, controlled, multicenter clinical trial in medication-free patients with PD/A who were treated with 12 sessions of manualized CBT, functional magnetic resonance imaging (fMRI) was used during fear conditioning before and after CBT. Quality-controlled fMRI data from 42 patients and 42 healthy subjects were obtained. Results: After CBT, patients compared to control subjects revealed reduced activation for the conditioned response (CS+ > CS–) in the left inferior frontal gyrus (IFG). This activation reduction was correlated with reduction in agoraphobic symptoms from t1 to t2. Patients compared to control subjects also demonstrated increased connectivity between the IFG and regions of the “fear network” (amygdalae, insulae, anterior cingulate cortex) across time. Conclusions: This study demonstrates the link between cerebral correlates of cognitive (IFG) and emotional (“fear network”) processing during symptom improvement across time in PD/A. Further research along this line has promising potential to support the development and further optimization of targeted treatments.

The temporal specificity of functional magnetic resonance imaging (fMRI) is limited by a sluggish and locally variable hemodynamic response trailing the neural activity by seconds. Here, we demonstrate for an attention capture paradigm that it is, never the less, possible to extract information about the relative timing of regional brain activity during cognitive processes on the scale of 100 ms by comparing alternative signal models representing early versus late activation. We demonstrate that model selection is not driven by confounding regional differences in hemodynamic delay. We show, including replication, that the activity in the dorsal anterior insula is an early signal predictive of behavioral performance, while amygdala and ventral anterior insula signals are not. This specific finding provides new insights into how the brain assigns salience to stimuli and emphasizes the role of the dorsal anterior insula in this context. The general analytic approach, named “Cognitive Timing through Model Comparison” (CTMC), offers an exciting and novel method to identify functional brain subunits and their causal interactions.

Gesicht und Stimme einer Person sind stark miteinander assoziiert und werden normalerweise als eine Einheit wahrgenommen. Trotz des natürlichen gemeinsamen Auftretens von Gesichtern und Stimmen, wurden deren Wahrnehmung in den Neurowissenschaften traditionell aus einer unisensorischen Perspektive untersucht. Das heißt, dass sich Forschung zu Gesichtswahrnehmung ausschließlich auf das visuelle System fokusierte, während Forschung zu Stimmwahrnehmung nur das auditorische System untersuchte. In dieser Arbeit schlage ich vor, dass das Gehirn an die multisensorische Beschaffenheit von Gesichtern und Stimmen adaptiert ist, und dass diese Adaption sogar dann sichtbar ist, wenn nur die Stimme einer Person gehört wird, ohne dass das Gesicht zu sehen ist. Im Besonderen, untersucht diese Arbeit wie das Gehirn zuvor gelernte Gesichts-Stimmassoziationen ausnutzt um die auditorische Analyse von Stimmen und Sprache zu optimieren. Diese Dissertation besteht aus drei empirischen Studien, welche raumzeitliche Hirnaktivität mittels funktionaler Magnetresonanztomographie (fMRT) und Magnetoenzephalographie (MEG) liefern. Alle Daten wurden gemessen, während Versuchspersonen auditive Sprachbeispiele von zuvor familiarisierten Sprechern (mit oder ohne Gesicht des Sprechers) hörten. Drei Ergebnisse zeigen, dass zuvor gelernte visuelle Sprecherinformationen zur auditorischen Analyse von Stimmen beitragen: (i) gesichtssensible Areale waren Teil des sensorischen Netzwerks, dass durch Stimmen aktiviert wurde, (ii) die auditorische Verarbeitung von Stimmen war durch die gelernte Gesichtsinformation zeitlich faszilitiert und (iii) multisensorische Interaktionen zwischen gesichtsensiblen und stimm-/sprachsensiblen Arealen waren verstärkt. Die vorliegende Arbeit stellt den traditionellen, unisensorischen Blickwinkel auf die Wahrnehmung von Stimmen und Sprache in Frage und legt nahe, dass die Wahrnehmung von Stimme und Sprache von von einem multisensorischen Verarbeitungsschema profitiert. / Face and voice of a person are strongly associated with each other and usually perceived as a single entity. Despite the natural co-occurrence of faces and voices, brain research has traditionally approached their perception from a unisensory perspective. This means that research into face perception has exclusively focused on the visual system, while research into voice perception has exclusively probed the auditory system. In this thesis, I suggest that the brain has adapted to the multisensory nature of faces and voices and that this adaptation is evident even when one input stream is missing, that is, when input is actually unisensory. Specifically, the current work investigates how the brain exploits previously learned voice-face associations to optimize the auditory processing of voices and vocal speech. Three empirical studies providing spatiotemporal brain data—via functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG)—constitute this thesis. All data were acquired while participants listened to auditory-only speech samples of previously familiarized speakers (with or without seeing the speakers’ faces). Three key findings demonstrate that previously learned visual speaker information support the auditory analysis of vocal sounds: (i) face-sensitive areas were part of the sensory network activated by voices, (ii) the auditory analysis of voices was temporally facilitated by learned facial associations and (iii) multisensory interactions between face- and voice/speech-sensitive regions were increased. The current work challenges traditional unisensory views on vocal perception and rather suggests that voice and vocal speech perception profit from a multisensory neural processing scheme.

Aufmerksamkeit auf visuelle Reize kann durch top-down Such- Strategien oder durch bottom-up Eigenschaften des visuellen Reizes gesteuert werden. Die Eigenschaft einer bestimmten Position, aus einer visuellen Szene heraus zu stechen, wird als Salienz bezeichnet. Es wird angenommen, dass auf neuronaler Ebene eine Salienzkarte existiert. Bis heute ist strittig, wo die Repräsentation einer solchen Karte im Gehirn lokalisiert sein könnte. Im Rahmen dieser Dissertation wurden drei Experimente durchgeführt, die verschiedene Aspekte von bottom-up Salienz-Verarbeitung mit Hilfe der funktionellen Magnetresonanztomographie untersuchten. Während die Aufmerksamkeit auf einen Fixationspunkt gerichtet war, wurde die neuronale Reaktion auf unterschiedlich saliente Stimuli in der Peripherie untersucht. In den ersten zwei Experimenten wurde die neuronale Antwort auf Orientierungskontrast und Luminanzkontrast untersucht. Die Ergebnisse deuten darauf hin, dass Salienz möglicherweise verteilt im visuellen System kodiert ist. Im dritten Experiment wurden natürliche Szenen als Stimuli verwendet. Im Einklang mit den Ergebnissen der ersten beiden Experimente wurde hier graduierte Salienz in frühen und späten visuellen Arealen identifiziert. Darüber hinaus konnten Informationen über die salientesten Positionen aus weiter anterior liegenden Arealen, wie dem anterioren intraparietalen Sulcus (aIPS) und dem frontalen Augenfeld (FAF), dekodiert werden. Zusammengenommen deuten die Ergebnisse auf eine verteilte Salienzverarbeitung von unterschiedlichen low-level Merkmalen in frühen und späten visuellen Arealen hin, die möglicherweise zu einer merkmalsunabhängigen Salienzrepräsentation im posterioren intraparetalen Sulcus zusammengefasst werden. Verschiebungen der Aufmerksamkeit zu den salientesten Positionen werden dann im aIPS und im FAF vorbereitet. Da die Probanden mit einer Fixationsaufgabe beschäftigt waren, wird die Salienz vermutlich automatisch verarbeitet. / Attention to visual stimuli can be guided by top-down search strategies or by bottom-up information. The property of a specific position to stand out in a visual scene is referred to as saliency. On the neural level, a representation of a saliency map is assumed to exist. However, to date it is still unclear where such a representation is located in the brain. This dissertation describes three experiments that investigated different aspects of bottom-up saliency processing in the human brain using functional magnetic resonance imaging (fMRI). Neural responses to different salient stimuli presented in the periphery were investigated while top-down attention was directed to the central fixation point. The first two experiments investigated the neural responses to orientation contrast and to luminance contrast. The results indicate that saliency is potentially encoded in a distributed fashion in the visual system and that a feature-independent saliency map is calculated late in the processing hierarchy. The third experiment used natural scenes as stimuli. Consistent with the results of the other two experiments, graded saliency was identified in striate and extrastriate visual cortex, in particular in posterior intraparietal sulcus (pIPS), potentially reflecting a representation of feature-independent saliency. Additionally information about the most salient positions could be decoded in more anterior brain regions, namely in anterior intraparietal sulcus (aIPS) and frontal eye fields (FEF). Taken together, the results suggest a distributed saliency processing of different low-level features in striate and extrastriate cortex that is potentially integrated to a feature-independent saliency representation in pIPS. Shifts of attention to the most salient positions are then prepared in aIPS and FEF. As participants were engaged in a fixation task, the saliency is presumably processed in an automatic manner.

In der realen Welt müssen Agenten oft aus teilweise-informativen Beobachtungen den (sich verändernden) latenten Zustand der Umgebung lernen, um optimale Entscheidungen zu treffen. Dies erfordert den dynamischen Wechsel zwischen relativer kognitiver Flexibilität und Stabilität. In dieser Dissertation stelle ich drei Studien vor, die untersuchten, ob zeitliche neuronale Variabilität, gemessen mit funktioneller Magnetresonanztomographie in Menschen, als neuronaler Marker der kognitiven Flexibilität in diesem Aufgabenkontext dient. Unsere Ergebnisse zeigten, dass bei jüngeren Erwachsenen effizienteres Lernen in einer stationären Umgebung mit einer stärkeren Kompression der neuronalen Variabilität einherging (Kapitel 1) und dass weniger neuronale Variabilität mit mehr kognitiver Stabilität und besserer Lernleistung in einer dynamischen Umgebung verbunden war (Kapitel 2). Bei älteren Erwachsenen hingegen war höhere neuronale Variabilität mit mehr kognitiver Stabilität verbunden, was eine Heuristik der Wiederholung vorheriger Entscheidungen darstellte (Kapitel 2). Zusammengenommen deuten diese Ergebnisse darauf hin, dass die Kompression von neuronaler Variabilität die Verwendung eines erlernten internen Modells der Umgebung reflektieren könnte, um Entscheidungen unter Unsicherheit zu treffen. Darüber hinaus fanden wir, dass individuelle Unterschiede in der Dopaminsystemkapazität, gemessen mit Positronen-Emissions-Tomographie, mit neuronaler Variabilität in verhaltensrelevanten Gehirnregionen zusammenhing (Kapitel 2), und dass die Verabreichung eines Dopamin-Vorläufers die neuronale Variabilität und Aufgabenleistung bei älteren Erwachsenen in einer dynamischen Lernumgebung gemeinsam modulierte (Kapitel 3), abhängig von der Reihenfolge der Medikamentengabe. Dies liefert erste Hinweise darauf, dass dopaminerge Mechanismen der neuronalen Variabilität während des Lernens und Entscheidungsfindung unter Unsicherheit über die menschliche Lebensspanne hinweg zugrunde liegt. / In the real world, agents often need to learn about the (changing) latent state of the environment from partially informative observations to make optimal decisions. This requires them to dynamically shift between mental modes of relative cognitive flexibility and stability. In this this dissertation I present three studies that investigated whether moment-to-moment neural variability, measured with functional magnetic resonance imaging in humans, may serve as a neural marker of cognitive flexibility in this task context. We found that in younger adults, more efficient learning in a stationary environment was related to more neural variability compression with increasing evidence exposure (Chapter 1) and that less neural variability related to more cognitive stability and better task performance during learning in a dynamic environment (Chapter 2). In older adults, however, more neural variability related to more cognitive stability, which reflected a decision heuristic to repeat previous choices (Chapter 2). Together, these findings suggest that neural variability compression may reflect the use of a learned internal model of the environment to guide decision-making under uncertainty, rather than reflecting cognitive flexibility/stability per se. Furthermore, we found that individual differences in dopamine system capacity, measured with positron emission tomography, was linked to neural variability in behaviorally-relevant brain regions (Chapter 2) and that administration of a dopamine precursor jointly modulated neural variability and task performance in older adults in a drug-order dependent manner (Chapter 3) during learning in a dynamic environment. These findings provide first evidence that dopaminergic mechanisms may support neural variability during learning and decision-making under uncertainty across the human lifespan.