Projects

Increased sensitivity to negative events and reduced responsivity to positive outcomes have been related to biased associative learning in major depressive disorder (MDD). We showed that uncontrollability, i.e. loss of behavioral control over aversive stimuli negatively affects subsequent operant learning in coping with stressors especially in depressed subjects. In parallel, we identified neural alterations in respondent learning during both expectation and delivery of reward and punishment in MDD. However, the interplay between altered operant and respondent learning in MDD has not been addressed in neuroimaging studies so far; although theories on learned helplessness and cognitive reactivity converge in the assumption that maladaptive information processing in MDD emerges during the confrontation with aversive stimuli. Furthermore, whether alterations in associative learning may represent state-dependent epiphenomena or stable vulnerability characteristics of the disorder is an open question, which can best be addressed by investigating both acutely and remitted depressed individuals. Therefore, we aim to implement a combined operant-respondent learning paradigm to investigate whether currently depressed (n=40) and currently remitted depressed patients (n=40) differ from healthy controls (n=60) in appetitive and aversive respondent learning during operantly conditioned states of controllability vs. uncontrollability. Specifically, we expect that both acutely depressed and remitted depressed individuals will show reduced positive reinforcement learning during the induction of uncontrollability due to their proposed vulnerability towards aversive stress-induced cognitive states. To assess the specificity of altered associative learning in depression, we will additionally investigate patients with anxiety disorders without lifetime history of depression (n=40) with the same paradigm. Participants will be subjected to a differential operant-respondent learning paradigm during 3 Tesla functional magnetic resonance imaging (fMRI), accompanied by peripheral physiological measures. Symbols will serve as conditioned stimuli which are differentially associated with appetitive and aversive reinforcement, and an auditory cue will act as unconditioned stimulus in a choice reaction task. Controllability status will be induced by perceived escapable and non-escapable electric stimulation in intermixed trials followed by conditioning trials. In addition to the identification of regional neural correlates of altered learning in depression and anxiety, fMRI will include the assessment of default network activities to target global maladaptive changes in neural networks subserving associative learning. Furthermore, we will investigate associations with genetic, endocrinological and cognitive vulnerability variables. Psychometric assessment will be in continuance of our previous work in the SFB 636. In addition, our samples will participate in a cooperation project with projects C06 and D06 in which a reward paradigm will be applied to assess common and specific alterations in reward learning across different disorders.